WO2019085760A1 - 一种图像处理方法及装置 - Google Patents
一种图像处理方法及装置 Download PDFInfo
- Publication number
- WO2019085760A1 WO2019085760A1 PCT/CN2018/110709 CN2018110709W WO2019085760A1 WO 2019085760 A1 WO2019085760 A1 WO 2019085760A1 CN 2018110709 W CN2018110709 W CN 2018110709W WO 2019085760 A1 WO2019085760 A1 WO 2019085760A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image data
- block
- image
- data block
- dimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/564—Depth or shape recovery from multiple images from contours
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61C—DENTISTRY; APPARATUS OR METHODS FOR ORAL OR DENTAL HYGIENE
- A61C9/00—Impression cups, i.e. impression trays; Impression methods
- A61C9/004—Means or methods for taking digitized impressions
- A61C9/0046—Data acquisition means or methods
- A61C9/0053—Optical means or methods, e.g. scanning the teeth by a laser or light beam
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10068—Endoscopic image
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30036—Dental; Teeth
Definitions
- the present invention relates to the field of image processing technologies, and in particular, to an image processing method and apparatus.
- the traditional oral endoscope is a kind of equipment used for dental optical modulating.
- the optical gun head for taking the mold needs to be moved in order on the user's upper dentition and lower dentition, and the scanning gun head is not supported in the oral cavity. Roaming freely, professionalism is very strong, user interaction is poor, and requires professional personnel to operate.
- the embodiment of the invention provides an image processing method and device, which solves the problem that the user interaction effect of the oral image presentation in the prior art is poor, and it is difficult to support the user self-service three-dimensional true color modulo.
- An image processing method comprising:
- Step A acquiring image data of a user collected by the endoscope; wherein the image data includes at least image data captured by the camera unit in the endoscope;
- Step B splicing the image data, and/or the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data;
- Step C respectively matching the image data block including the image data in the spliced image data block with the image of the block in the saved three-dimensional image frame database, wherein the three-dimensional image frame database stores three-dimensional image frames The data of the image of the block divided by the image frame image and the position information of the image of each block;
- Step D if it is determined that at least one mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is obtained, determining the at least one image including the image data according to the mapping relationship Positioning a block corresponding to the data block in the saved three-dimensional image contour of the user, and reconstructing the at least one image data block containing the image data in a corresponding determined position in the three-dimensional image contour of the user Obtaining the reconstructed three-dimensional image data, updating the currently saved three-dimensional image model of the user according to the reconstructed three-dimensional image data, and updating the user according to the first preset display policy And displaying, in the spliced image data block, all or part of other image data blocks that have been determined to obtain a mapping relationship; wherein an initial value of the user's three-dimensional image model is the three-dimensionality of the user Image outline
- Step E if it is determined that the preset condition is met, the updated three-dimensional image model of the user and the recorded number information of the block that has been determined to obtain the mapping relationship, and the stitched according to a preset rule Part or all of the image data blocks in the image data block except for the image data blocks that have been determined to obtain the mapping relationship with the blocks in the three-dimensional image frame database are uploaded to the cloud server, so that the cloud server according to the updated a three-dimensional image model of the user and number information of the block that has been determined to obtain the mapping relationship, and an image data block in the stitched image data block except that the mapping relationship with the block in the three-dimensional image frame database has been determined
- the saved partial image database database is updated by part or all of the image data blocks.
- the location information of the image of the block includes: a spatial positional relationship between each block;
- An image of each of the three-dimensional image contours is a three-dimensional curved shape based on an image of the block in the three-dimensional image frame database or the three-dimensional image model of the user, including a preset single color and a single texture image .
- the method further includes:
- the updated three-dimensional image model of the user and all or part of the other image data blocks in the stitched image data block that have been determined to obtain a mapping relationship are displayed. Specifically, including:
- Arranging, according to the area of the image data block or the image data block, the number of the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship are sorted from large to small;
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data blocks, and the updated three-dimensional image model of the user is sequentially displayed in the d display regions. And the first d-1 image data blocks taken out;
- n is less than d-1
- the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as that of n+1, and the update is displayed in the display area in the adjusted current display interface.
- the method further comprises:
- the number information of the block that has determined to obtain the mapping relationship is recorded according to the mapping relationship.
- the step D further includes:
- the currently saved three-dimensional image model of the user and all or part of the other image data blocks in the stitched image data block that have been determined to obtain a mapping relationship are displayed. Specifically, including:
- Arranging, according to the area of the image data block or the image data block, the number of the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship are sorted from large to small;
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data block, and the currently saved three-dimensional image model of the user is sequentially displayed in the d display regions. And the first d-1 image data blocks taken out;
- n is less than d-1, the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as that of n+1, and the current display areas in the adjusted current display interface are sequentially displayed.
- the step D further includes:
- all or part of the spliced image data block is displayed according to the second preset display strategy, and specifically includes:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block;
- the method further includes:
- the method further comprises:
- An image processing method comprising:
- Step A acquiring image data of a user collected by the endoscope; wherein the image data includes at least image data captured by the camera unit in the endoscope;
- Step B splicing the image data and the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data;
- Step C respectively matching the image data block including the image data in the spliced image data block with the image of the block in the saved three-dimensional image frame database, wherein the three-dimensional image frame database stores three-dimensional image frames The data of the image of the block divided by the image frame image and the position information of the image of each block;
- Step D if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the spliced image data block is not obtained in the three-dimensional image frame database
- the image data block of the mapping relationship of the block is displayed on all or part of the spliced image data block according to the second preset display strategy;
- Step E If it is determined that the preset condition is met, part or all of the spliced image data block is uploaded to the cloud server according to a preset rule, so that the cloud server is configured according to the spliced image data block. Part or all, the saved three-dimensional image frame database is updated.
- all or part of the spliced image data block is displayed according to the second preset display strategy, and specifically includes:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block;
- the first d image data blocks are taken out from the sorted image data blocks, and respectively In the d display areas, the first d image data blocks are sequentially displayed, and if a is smaller than d, the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as a, and Displaying the a plurality of image data blocks in sequence in the display area in the adjusted current display interface;
- the method further includes, after performing step D, returning to performing step A.
- the method further includes:
- An image processing apparatus comprising:
- a receiving unit configured to acquire image data of a user collected by the endoscope; wherein the image data includes at least image data captured by the camera unit in the endoscope;
- a processing unit configured to splicing the image data and the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data; and respectively And the image data block including the image data in the spliced image data block is matched with the image of the block in the saved three-dimensional image frame database, and if it is determined that at least one image data block including the image data is obtained and the three-dimensional image Determining a mapping relationship of the blocks in the image frame database, determining, according to the mapping relationship, a position of the at least one block corresponding to the image data block including the image data in a saved three-dimensional image contour of the user, and The at least one image data block including the image data is reconstructed at a corresponding determined position in the three-dimensional image contour of the user, and the reconstructed three-dimensional image data is obtained, according to the reconstructed three-dimensional image data, Updating a currently saved three-dimensional image model of the user; wherein the three-dimensional image
- a display unit configured to perform, according to the first preset display policy, all or part of the updated three-dimensional image model of the user and other image data blocks of the stitched image data block that have been determined to obtain a mapping relationship Display
- a sending unit configured to: if it is determined that the preset condition is met, the updated three-dimensional image model of the user and the recorded number information of the block that has been determined to obtain the mapping relationship, and the splicing according to a preset rule And all or part of the image data block except the image data block that has been determined to obtain the mapping relationship with the block in the three-dimensional image frame database is uploaded to the cloud server, so that the cloud server is updated according to the updated a three-dimensional image model of the user and number information of the block that has been determined to obtain the mapping relationship, and image data of the stitched image data block except that the mapping relationship with the block in the three-dimensional image frame database has been determined
- the saved three-dimensional image frame database is updated by part or all of the image data blocks outside the block.
- the location information of the image of the block includes: a spatial positional relationship between each block;
- An image of each of the three-dimensional image contours is a three-dimensional curved shape based on an image of the block in the three-dimensional image frame database or the three-dimensional image model of the user, including a preset single color and a single texture image .
- the processing unit is further configured to:
- the updated three-dimensional image model of the user and all or part of the other image data blocks in the stitched image data block that have been determined to obtain a mapping relationship are displayed.
- the display unit is specifically used to:
- Arranging, according to the area of the image data block or the image data block, the number of the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship are sorted from large to small;
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data blocks, and the updated three-dimensional image model of the user is sequentially displayed in the d display regions. And the first d-1 image data blocks taken out;
- n is less than d-1
- the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as that of n+1, and the update is displayed in the display area in the adjusted current display interface.
- the processing unit is further configured to:
- the number information of the block that has determined to obtain the mapping relationship is recorded according to the mapping relationship.
- the processing unit is further configured to: if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and at least one of the spliced image data blocks is Obtaining an image data block that has a mapping relationship with a block in the three-dimensional image frame database, the display unit is further configured to:
- the display unit is specifically used to:
- Arranging, according to the area of the image data block or the image data block, the number of the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship are sorted from large to small;
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data block, and the currently saved three-dimensional image model of the user is sequentially displayed in the d display regions. And the first d-1 image data blocks taken out;
- n is less than d-1, the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as that of n+1, and the current display areas in the adjusted current display interface are sequentially displayed.
- the processing unit is further configured to: if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the obtained image data block is not obtained in the spliced image data block
- the image data block of the mapping relationship of the blocks in the three-dimensional image framework database is further used for:
- all or part of the spliced image data block is displayed according to the second preset display strategy, and the display unit is specifically configured to:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block;
- the processing unit is further configured to:
- the receiving unit is further configured to: in the display unit, determine, according to the first preset display policy, that the updated three-dimensional image model of the user and the stitched image data block are determined to obtain a mapping. After all or part of the other image data blocks of the relationship are displayed, the image data of the user acquired by the acquisition of the endoscope is returned.
- An image processing apparatus comprising:
- a receiving unit configured to acquire image data of a user collected by the endoscope; wherein the image data includes at least image data captured by the camera unit in the endoscope;
- a processing unit configured to splicing the image data and the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data; and respectively The image data block including the image data in the spliced image data block is matched with the image of the block in the saved three-dimensional image frame database; wherein the three-dimensional image frame database stores a region that divides the three-dimensional image frame image The data of the image of the block and the position information of the image of each block;
- a display unit configured to: if the processing unit determines that a mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the three-dimensional image data block is not obtained and the three-dimensional The image data block of the mapping relationship of the blocks in the image frame database is displayed on all or part of the spliced image data block according to the second preset display strategy;
- a sending unit if it is determined that the preset condition is met, uploading part or all of the spliced image data block to the cloud server according to a preset rule, so that the cloud server is configured according to the spliced image data Part or all of the block, the saved three-dimensional image frame database is updated.
- all or part of the spliced image data block is displayed according to the second preset display strategy, and the display unit is specifically configured to:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block;
- the first d image data blocks are taken out from the sorted image data blocks, and respectively In the d display areas, the first d image data blocks are sequentially displayed, and if a is smaller than d, the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as a, and Displaying the a plurality of image data blocks in sequence in the display area in the adjusted current display interface;
- the receiving unit is further configured to: after the displaying unit displays all or part of the spliced image data block according to the second preset display policy, returning to performing image data of the user acquired by the endoscope ;
- the processing unit is further configured to:
- the image data of the user collected by the endoscope is acquired; wherein the image data includes at least image data captured by the camera unit in the endoscope, and the type of the image data is a depth image. And the image data of the collected user indicates that the splicing process is not performed; splicing the image data and the image data and the saved image data block to obtain the spliced image data block, wherein the image data is The block includes at least one image data; the image data block including the image data in the stitched image data block is respectively matched with the image of the block in the saved three-dimensional image frame database, wherein the three-dimensional image frame database Storing data of an image of a block in which the three-dimensional image frame image is divided and position information of an image of each block; the image data of the block includes: number information, image feature information; and performing according to different matching results Corresponding different presentation methods and different information uploaded to the cloud server to make the cloud service Updating the three-dimensional image frame database, so that the acquired image data
- the spliced image data block is displayed, and the endoscope is not required to continuously scan the oral cavity.
- the user can use the endoscope to scan the oral cavity at will, as long as the image data of the inner surface of the oral cavity is obtained regardless of order, and whether or not it matches. If you are successful, you can display the three-dimensional image of the oral cavity. It does not require professional operation. It supports the user's self-service oral endoscopy, and can not only present the three-dimensional image of the user's mouth, but also can display it dynamically, and the display effect is better and improved. User experience and interaction effects, Good self-supporting three-dimensional true color user modulo.
- the cloud server may update the three-dimensional image frame database according to the information, and solve the problem that the three-dimensional image processing of the oral cavity performs conditional matching and display.
- the 3D image framework database can be continuously updated to gradually enrich and improve the 3D image framework database, the process is more complete, and it is closer to the actual engineering implementation, and then, with continuous use and update, the 3D image framework database can be improved later.
- the success rate of the matching can make the three-dimensional true color image contained in the user's three-dimensional image model more and more, and the display effect is improved.
- FIG. 1 is a flowchart of an image processing method according to Embodiment 1 of the present invention.
- FIG. 2 is a schematic diagram of partitioning of a display area in a current interface according to Embodiment 1 of the present invention
- FIG. 3 is a flowchart of a method for updating a cloud server according to Embodiment 2 of the present invention.
- FIG. 5 is a flowchart of a method for updating a cloud server according to Embodiment 4 of the present invention.
- FIG. 6 is a schematic diagram showing a three-dimensional image display effect according to Embodiment 9 of the present invention.
- FIG. 7 is a schematic diagram showing a three-dimensional image display effect according to Embodiment 10 of the present invention.
- FIG. 8 is a schematic diagram showing a three-dimensional image display effect according to Embodiment 11 of the present invention.
- FIG. 9 is a schematic diagram showing a three-dimensional image display effect according to Embodiment 13 of the present invention.
- FIG. 10 is a schematic structural diagram of an image processing apparatus according to Embodiment 15 of the present invention.
- FIG. 11 is a schematic structural diagram of an image processing apparatus according to Embodiment 16 of the present invention.
- the three-dimensional image is mainly for the oral cavity, wherein the endoscope is an oral endoscope, and of course, the embodiment of the present invention is not limited to the three-dimensional image of the oral cavity, and the three-dimensional image for other fields. The same can be applied.
- the following is only an example of the oral cavity.
- Embodiment 1 is a diagrammatic representation of Embodiment 1:
- the specific process of the image processing method is as follows:
- Step 100 Acquire image data of a user collected by an endoscope; wherein the image data includes at least image data captured by an imaging unit in the endoscope, and the type of the image data is a depth image; The image data of the collected user indicates that the stitching process has not been performed.
- the endoscope is, for example, an oral endoscope, and the endoscope is provided with an imaging unit for capturing an image.
- the endoscope may be provided with an imaging unit, or may be provided with a plurality of imaging units, and may be photographed.
- the type of image data that arrives is a depth image, that is, an RGBD image, which is a three-dimensional true color image, so that three-dimensional information of the image can be acquired, which facilitates subsequent three-dimensional image processing.
- the oral endoscope of the embodiment of the present invention can be provided with two imaging units back to back on the endoscope head, and the two imaging units are relatively fixed in position and pre-calibrated, synchronized, and synchronized.
- the image data synchronously captured by the two camera units respectively have corresponding camera unit identifiers, such as number information of the camera unit, and have the same timing stamp information, wherein the time stamp information indicates that the power is turned on from the endoscope.
- the image data is collected several times, so that the time stamp can be set for each acquired image data, which is also convenient for distinguishing image data collected at different times.
- the image data acquired in step 100 is a plurality of image data without splicing processing with camera unit identification and timing stamp information, and the image data may have the same timing stamp or may have different timing stamps. This depends on the number of camera units in the endoscope, that is, the number of image data acquired by the endoscope every time, and the number of image data that is set each time the processing is performed.
- the method further includes: determining that the number of image data is greater than a preset threshold.
- the endoscope acquires two pieces of image data each time, and the preset threshold is 6, that is, the number of image data that is set to be processed each time is six, and the user who acquires the endoscope 3 shots needs to be acquired.
- the image data is subjected to subsequent stitching processing.
- Image data are image data P(1), P(2), P(3), P(4), P(5), P(6, respectively.
- Step 110 splicing the image data, and/or the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data.
- the image data and the image data and the saved image data block are respectively judged and spliced with each other, and the spliced image data block is obtained, saved, and updated according to the spliced image data block.
- the saved image data block that is, in the embodiment of the present invention, the spliced image data block represents all the spliced processed image data that can be spliced and cannot be spliced, wherein not only the splicing process but also the splicing is successful,
- the larger-area image data composed of the other image data is spliced, and the image data that is still spliced unsuccessfully after being judged is still isolated, and includes not only the image data acquired in step 100 but the image data to be processed.
- the image data block further includes an image data block that does not include the image data acquired in step 100.
- the image data is transformed into image data blocks after being spliced, and the spliced image data block may include an image data, indicating that the image data block cannot be included.
- Spliced single image data which may also include multiple image data, a table The larger image data obtained by splicing and splicing the plurality of image data is shown.
- the number of image data included in the spliced image data block may also be recorded, so as to facilitate subsequent sorting of the image data blocks.
- the acquired image data is first spliced, and subsequent matching is performed, and the spliced image data block area is relatively large, which can improve the matching efficiency, reduce the number of images, and reduce the image and area. The number of times the block is matched, the time is reduced, and the execution efficiency is also improved.
- Step 120 Match the image data block including the image data in the spliced image data block with the image of the block in the saved three-dimensional image frame database, where the three-dimensional image frame database stores three-dimensional image frames.
- a three-dimensional image frame database and a three-dimensional image outline of the oral cavity are established in advance.
- a three-dimensional image frame database is constructed based on various situations of a human oral cavity, and the three-dimensional image frame database stores general frame data of a three-dimensional image model of a human oral cavity, and the frame data covers various situations.
- Image feature information of all surface areas of the human oral cavity such as shape features, color features, texture features, and the like.
- the three-dimensional image frame database stores image data of the block that divides the three-dimensional image frame image and position information of the image of each block; the position information of the image of the block includes: spatial positional relationship between each block
- the image data of the block includes: number information, image feature information.
- the three-dimensional image outline stores shape contour data of a three-dimensional image of each region (including each block) of the inner surface of the human full oral cavity.
- the three-dimensional image contour of the user stores at least shape contour data of the three-dimensional image of each block in the oral cavity of the user.
- the three-dimensional image contour of the user may be updated according to the actual situation of each user. With the use of different users, for each user, there will be a three-dimensional image belonging to one's own mouth. Image outline.
- An image of each of the three-dimensional image contours is a three-dimensional curved shape based on an image of the block in the three-dimensional image frame database or the three-dimensional image model of the user, including a preset single color and a single texture image .
- the method includes: determining, according to the image feature information of the block, the spatial positional relationship between the blocks, and/or the number information, or the identifier of the camera unit carried according to the image data in the image data block.
- the relative spatial positional relationship of the subsequent image data blocks respectively matches the image data blocks containing the image data in the stitched image data block with the images of the blocks in the saved three-dimensional image frame database.
- the image data block including the image data in step 100 in the spliced image data block is matched with the image of the block in the three-dimensional image frame database, wherein the image data in step 100 is matched. That is, the image data to be processed this time, because the currently saved image data block must be subjected to the matching processing before the splicing process, and in the embodiment of the present invention, if the image data block is successfully matched. It will be reconstructed in the user's 3D image contour, so the currently saved image data blocks are unsuccessful. If the stitching process is performed this time, all the stitched image data blocks are matched, for those and new ones. The image data cannot be spliced, and the image data block is still saved. The matching is also unsuccessful. The matching again increases the number of matching and reduces the processing efficiency. Therefore, in the embodiment of the present invention, the image data is Blocks do not need to be matched again, saving time and increasing efficiency.
- the image data of the user acquired by the endoscope is P(1) and P(2)
- P(1), P(2) and P(1), P(2) and Q'(1),..., Q'(m),...,Q'(32 Splicing if the image data P(1) can be spliced with the first image data block Q'(1) stored in the system, and can be spliced with the fourth image data block Q'(4) stored in the system, If the image data P(2) cannot be spliced with any other image data block, the image data P(1), the image data block Q'(1), and the image data block Q'(4) are spliced to form new image data.
- the block, denoted as Q' (Temp 1), and the image data P(2) also constitute an independent new image data block, denoted as Q'(Temp 2), then only Q' (Temp 1) when matching And Q' (Temp 2) matches the image of the block in the 3D image framework database.
- the method further includes:
- image pattern recognition is used to determine whether the endoscope has been placed in the oral cavity, and the user does not need to manually open the oral endoscope through the switch.
- the endoscope does not need to be provided with the switch, thereby reducing the hardware. Settings also improve the user experience.
- a switch is set on the endoscope, which is manually controlled by the user. After the user puts the endoscope into the oral cavity and manually closes the switch, the system considers the inside The speculum has been placed in the mouth.
- the three-dimensional image processing is mainly for the oral cavity. Therefore, according to the image pattern recognition algorithm, if it is determined that the image data block is not an oral image, the subsequent matching process may not be performed to avoid invalid operation. Determining that the image data block is an oral image, it can be determined that the endoscope has been placed in the oral cavity, and the subsequent matching operation can be continued.
- Step 130 If it is determined that a mapping relationship between at least one image data block including the image data and a block in the three-dimensional image frame database is obtained, determining, according to the mapping relationship, the at least one image including the image data Positioning a block corresponding to the data block in the saved three-dimensional image contour of the user, and reconstructing the at least one image data block containing the image data in a corresponding determined position in the three-dimensional image contour of the user Obtaining the reconstructed three-dimensional image data, updating the currently saved three-dimensional image model of the user according to the reconstructed three-dimensional image data, and updating the user according to the first preset display policy And displaying, in the spliced image data block, all or part of other image data blocks that have been determined to obtain a mapping relationship; wherein an initial value of the user's three-dimensional image model is the three-dimensionality of the user Image outline.
- step 130 the method specifically includes:
- the initial value of the three-dimensional image model of the user is a three-dimensional image contour of the user.
- the method includes: 1) determining, according to the mapping relationship, a location of the at least one block corresponding to the image data block including the image data in a saved three-dimensional image contour of the user.
- the position in the three-dimensional image contour of the user may be determined in the manner of the spatial positional relationship and/or the numbering information of the blocks, the spatial positional relationship between the image data blocks, and the like. There are no restrictions.
- the three-dimensional curved surface image belonging to the corresponding block is extracted from the at least one image data block including the image data according to the boundary feature information of the block in the three-dimensional image frame database; wherein the image feature information is At least the boundary feature information of the block is included; the extracted three-dimensional surface image is replaced with the image at the corresponding determined position in the three-dimensional image contour of the user, and the reconstructed three-dimensional image data is obtained.
- the reconstructed three-dimensional image data is replaced with an image at a corresponding determined position in the currently saved three-dimensional image model of the user.
- the image at the corresponding position in the user's three-dimensional image model can be continuously replaced, thereby realizing the effect of dynamically updating the user's three-dimensional image model.
- the user's three-dimensional image contour can also be updated, specifically:
- the three-dimensional image framework database initially established may not be comprehensive and cannot contain the oral image in all cases, so that the image data block and the three-dimensional image frame database are in the block.
- the problem is mainly solved for the establishment of the three-dimensional image framework database in the actual situation, and the conditional execution block matching and display in the oral three-dimensional image processing is solved.
- the problem is how to continuously update and improve the 3D image framework database, and how to display and process when the matching is successful and unsuccessful, so that the 3D image framework database can be gradually enriched and improved, and the process is more complete and closer to reality.
- the three-dimensional image model of the user may be updated to implement the subsequent three-dimensional image model display.
- the number information of the block that has determined to obtain the mapping relationship is recorded according to the mapping relationship.
- the number information of the block that has been determined to obtain the mapping relationship is recorded for later updating the three-dimensional image frame database, and the number or position of the block in the three-dimensional image frame database or the image data of the block may be used.
- the image feature information of the block also includes information such as adding a block level operation such as a block or a block to update.
- the method includes: 1) sorting, according to the area of the image data block or the number of image data blocks including the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship.
- the image data block is composed of one or more image data. Therefore, generally, the more the image data block includes the number of image data, the larger the area of the image data block, and may of course include a smaller number. The area of the image data block is rather larger, which has no effect on the subsequent sorting and display.
- the number of image data included may be sorted according to the number of image data included, or may be sorted according to the area thereof. Restrictions can be sorted in one way.
- the collected image data is a depth image and carries the image capturing unit identifier. Therefore, the area of the image data block can be calculated according to the internal calibration parameter and the depth image data of the imaging unit, and specifically calculated.
- the method of the area of the image data block can be based on the method in the prior art, and is not limited in the embodiment of the present invention, and will not be described in detail herein.
- the spliced image data blocks are determined to be obtained.
- the other image data blocks of the mapping relationship are Q'(1), ..., Q'(n) after being sorted from large to small according to the number of pieces of image data.
- the purpose of sorting is to preferentially display image data blocks that are relatively large in area or include a large number of image data, because the area is larger or the number of image data included is larger. The more image information is reflected, the more convenient the user can view and understand the displayed image.
- n and d-1 are judged because the image data block and the block have been successfully matched at this time, and the user's three-dimensional image model can be updated, and an area needs to be reserved for displaying the user's three-dimensional image during display.
- the model therefore, has d-1 for the display area of the image data block that did not match successfully.
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data blocks, and the updated three-dimensionality of the user is sequentially displayed in the d display areas.
- the image model and the extracted first d-1 image data blocks are discarded.
- n is less than d-1, adjust the number of display areas in the current display interface, so that the number of display areas is the same as n+1, and respectively in the display area in the adjusted current display interface, The updated three-dimensional image model of the user and the sorted image data block are displayed.
- n is 2 and d is 4, if 2 ⁇ (4-1) is judged, the number of current display areas is relatively large, and there will be no blank area for displaying images, so as to further enhance the user's viewing experience and enhance the display effect. Therefore, it is necessary to reduce the number of display areas, reduce d to 3, and change to 3 display areas in the current display interface, and display the updated user's 3D image model in the first display area during display.
- Q'(1) is displayed in the second display area and Q'(2) is displayed in the third display area.
- step 130 when performing step 130, in addition to the above, if it is determined that at least one mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is obtained, the following Situation:
- a first case if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and at least one of the spliced image data blocks has been obtained and The image data block of the mapping relationship of the blocks in the three-dimensional image frame database is determined according to the first preset display strategy, and the currently saved three-dimensional image model of the user and the stitched image data block are determined to be obtained. All or part of the other image data blocks of the relationship are presented.
- the oral endoscope since the oral endoscope is turned on, and the endoscopic operation is started using the oral endoscope, the image data collected by the oral endoscope is continuously acquired, and the number of image data reaches a preset threshold every time. Then, a splicing and matching operation is started until the end of the endoscopic operation.
- the first case here is that after the splicing and matching processing is performed, the image data block including the image data of the current processing is not obtained and the three-dimensional image.
- the mapping relationship of the blocks in the framework database but since the oral endoscope is turned on, the endoscopic operation is started using the oral endoscope, and the 3D image has been obtained in a certain process before the splicing and matching processing.
- the method includes: 1) sorting, according to the area of the image data block or the number of image data blocks including the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship.
- n is not less than d-1
- the first d-1 image data blocks are taken out from the sorted image data blocks, and the currently saved three-dimensional user is sequentially displayed in the d display areas.
- n is less than d-1, adjust the number of display areas in the current display interface, so that the number of display areas is the same as n+1, and respectively in the display area in the adjusted current display interface, The currently saved three-dimensional image model of the user and the sorted image data block are displayed.
- the first preset display strategy in the first case is the same as the first preset display strategy in step 130, except that the displayed content is different, and therefore, it will not be described in detail here.
- a second case if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the three-dimensional image frame is not obtained in the spliced image data block
- the image data block of the mapping relationship of the blocks in the database is displayed on all or part of the spliced image data block according to the second preset display strategy.
- the second case here is for the case where the oral endoscope is turned on and the endoscopic operation is started using the oral endoscope, and the mapping relationship with the block in the three-dimensional image frame database has not been obtained.
- the spliced image data block is a plurality of mutually independent image data blocks. This is because, in the embodiment of the present invention, the endoscope is provided with two or more camera units to acquire images synchronously. Therefore, the image processing method in the embodiment of the present invention supports the user to complete the oral endoscopic operation by itself, without the user having Order, continuous scanning.
- the method includes: 1) if the number of the spliced image data blocks is a, the spliced image data blocks are changed from large to small according to an area of the image data block or an image data block including the number of image data. put in order.
- step 130 the above step 100 is performed.
- the image data collected by the endoscope is continuously acquired, and the current display is updated through continuous stitching and matching.
- the image data block and/or the user's three-dimensional image model enhances the display effect, and the user can see the user's three-dimensional image model.
- the user can see the displayed spliced image data block.
- Step 140 If it is determined that the preset condition is met, the updated three-dimensional image model of the user and the recorded number information of the block that has been determined to obtain the mapping relationship, and the stitched according to a preset rule Part or all of the image data blocks in the image data block except for the image data blocks that have been determined to obtain the mapping relationship with the blocks in the three-dimensional image frame database are uploaded to the cloud server, so that the cloud server according to the updated a three-dimensional image model of the user and number information of the block that has been determined to obtain the mapping relationship, and an image data block in the stitched image data block except that the mapping relationship with the block in the three-dimensional image frame database has been determined
- the saved partial image database database is updated by part or all of the image data blocks.
- the preset condition may be the following: determining that the endoscope is turned off, or the area of the reconstructed area in the three-dimensional image model of the user is greater than a preset threshold, or the area of the spliced image data block reaches a preset threshold.
- the limitation is not limited. Therefore, in the embodiment of the present invention, not every time the user's three-dimensional image model is updated and the new number information of the block that has been determined to obtain the mapping relationship is uploaded, However, when the conditions are met, the uploading is unified, which not only reduces the number of uploads, reduces server pressure, but also improves the efficiency of the cloud server when updating the 3D image framework database.
- the number information of the block that has been determined to obtain the mapping relationship that is uploaded to the cloud server includes not only the number information of the block that obtains the mapping relationship after the splicing and matching processing, but also includes the start of the endoscopic operation.
- the splicing and matching process records all the number information of the blocks that have been determined to obtain the mapping relationship.
- the predetermined rule for example, the spliced image data block includes the number of image data not less than a preset number, for example, five, or the area of the spliced image data block is not less than a preset area, for example, 100 mm. 2 , the embodiment of the present invention is not limited, and may be set according to actual conditions.
- the information uploaded to the cloud server may include the following parts: 1) a three-dimensional image model of the user. 2) Recorded number information of the block in which the mapping relationship with the spliced image data block is determined. 3) Part or all of the image data blocks in the spliced image data block except for the image data blocks that have been determined to obtain a mapping relationship with the blocks in the three-dimensional image frame database.
- the cloud server can continuously update the three-dimensional image framework database according to the information, specifically updating the number information of the block, increasing the block, and increasing the general frame data of the three-dimensional image model of the human mouth in different situations, and gradually updating and perfecting.
- the 3D image framework database forms a database of powerful and complete oral 3D image frames, with a complete labeling system that improves the success rate of database matching after using the 3D image framework.
- the information uploaded to the cloud server at different stages in the actual implementation process may be different.
- the actual implementation process may be divided into three phases:
- Stage 1 In the early stage, because the 3D image frame database is still not perfect, it is possible that the image data blocks after splicing cannot be successfully matched with the image of the block in the 3D image frame database, and the image data blocks after splicing cannot be obtained.
- the block of the mapping relationship therefore, may not be able to obtain the above-mentioned first part and two parts of information, that is, 1) the user's three-dimensional image model and 2) the recorded number information of the block that has been determined to obtain the mapping relationship with the spliced image data block.
- the information uploaded to the cloud server may only be the third part of the information, that is, 3) the image data block in the spliced image data block except the image data block that has been determined to obtain the mapping relationship with the saved block in the three-dimensional image frame database. Part or all of the image data block, and then the cloud server can only be based on the portion of the spliced image data block other than the image data block that has been determined to obtain the mapping relationship with the saved block in the three-dimensional image frame database or Update the 3D image frame database for all image data blocks.
- Phase 2 As the 3D image framework database is continuously used and updated, the matching success rate is improved, and at least one image data block can be successfully matched with the image of the block in the 3D image framework database, thereby updating the user's 3D image module, and The number information of the block that has been determined to obtain the mapping relationship with the spliced image data block can be obtained.
- the information uploaded to the cloud server at this time will be Including: 1) The user's 3D image model. 2) Recorded number information of the block in which the mapping relationship with the spliced image data block is determined. 3) part or all of the image data blocks except the image data block that has been determined to obtain the mapping relationship with the block in the saved three-dimensional image frame database, and then the cloud server can be based on the three parts. Information, update the 3D image framework database.
- Stage 3 After the image processing method of the embodiment of the present invention is operated for a long time, the mature stage can be reached.
- the 3D image framework database is relatively complete, and can include images of all oral conditions and labels of the 3D image framework database.
- the system is also perfect.
- the label system includes at least the block number information, relative position information, image information, etc. Therefore, when matching, the success rate will be greatly improved, and there may be no unsuccessful matching. Therefore, At this time, it may not be obtained that 3) part or all of the image data blocks except the image data block that have been determined to obtain the mapping relationship with the saved block in the three-dimensional image frame database are uploaded to the cloud server.
- the information is only 1) the user's 3D image model and 2) the recorded number information of the block that has been determined to obtain the mapping relationship with the stitched image data block, and then the cloud server according to the user's three-dimensional image model and record The number information of the block that has been determined to obtain the mapping relationship with the spliced image data block, Three-dimensional image frame database.
- the image data of the user collected by the acquired endoscope and the image data are spliced with the saved image data block to obtain the spliced image data block, and then the saved three-dimensional image frame database
- the image of the middle block is matched, and the user's three-dimensional image model and the stitched image data block are displayed according to different matching results and corresponding different display strategies. Finally, if the predetermined condition is met, the user's three-dimensional image model is adopted.
- the cloud server can continuously update the three-dimensional image frame database, and solve the problem of conditional execution block matching and display in the three-dimensional image processing of the oral cavity, which can be continuously updated and Improve the 3D image framework database, and the matching success and failure Effectively display and process, thus gradually enrich and improve the 3D image framework database, improve the matching success rate, and then with the continuous improvement of the 3D image framework database, in the later matching, the basic can be successfully matched, and can continue to match
- the successful image data block is reconstructed into a corresponding determined position in the user's three-dimensional image contour, and the user's three-dimensional image model is updated, so that the displayed user's three-dimensional image model will contain more and more three-dimensional true color
- each image data block can be used as an initial region.
- users can Randomly scan any position in the oral cavity, you can start continuous scanning from these multiple initial areas, or you can scan from other positions to form a new initial area, without continuous scanning from the unique initial area in the oral cavity, which is convenient for users to use.
- the user experience is improved, and the three-dimensional image of the scanned oral cavity is dynamically displayed, and the rendering effect is better, more convenient and flexible.
- Embodiment 2 is a diagrammatic representation of Embodiment 1:
- Step 300 Receive an update request sent by the terminal, where the update request includes at least: a three-dimensional image model of the user, a recorded number information of the block that has been determined to obtain a mapping relationship with the spliced image data block, and splicing Some or all of the image data blocks other than the image data block in which the mapping relationship with the block in the saved three-dimensional image frame database has been determined is determined.
- the image data of the block in which the three-dimensional image frame image is divided and the position information of the image of each block are stored in the three-dimensional image frame database; the image data of the block includes: number information and image feature information;
- the spliced image data block is obtained after the splicing process is performed by the terminal according to the image data of the user collected by the endoscope.
- the update request may include different information, which is specifically described in the foregoing first embodiment, and will not be further described herein.
- Step 310 Update the saved three-dimensional image framework database according to the update request.
- the cloud server receives three kinds of information, that is, the user's three-dimensional image model, when the spliced image data block and the three-dimensional image frame database block are successfully matched or unmatched successfully.
- the portion of the three-dimensional true color curved surface that has been determined to obtain the mapping relationship with the stitched image data block is beyond the existing three-dimensional image frame database (now There is a label system) an image of a known block label.
- the processing method of the embodiment of the present invention fills the mapping between the determined and the image data blocks after the stitching by performing the conventional image stitching.
- the gaps thus satisfy the needs of the user's oral endoscopic and three-dimensional modeling.
- it is still required to transmit the three-dimensional image model of the user and the recorded number information of the block that has been determined to obtain the mapping relationship with the spliced image data block to the cloud server.
- the cloud server After the cloud server receives it, it is necessary to detect that there are other three-dimensional true color curved surface image portions that have not been recognized between the blocks in the three-dimensional image model of the user that have been determined to obtain the mapping relationship. Then, the cloud server distributes the user's three-dimensional image model of the other three-dimensional true color curved image portion that has not been recognized to the corresponding dental professional, so that the corresponding dental expert can view the other three-dimensional true color curved image portions that have not yet been recognized. Conduct research and analysis, and update the labeling system to update the latest 3D image framework database (oral endoscopic panoramic label system).
- the cloud server may also be preset according to the preset image data block. Strategy, distribute these image data blocks to the corresponding dental experts, so that the corresponding dental experts can research and analyze these image data blocks, and update the label system to update the latest 3D image frame database (oral endoscopic panoramic label system) .
- the cloud server can continuously update the three-dimensional image frame database, and solve the problem that the three-dimensional image processing of the oral cavity can perform condition matching and display, and can continuously update and improve the three-dimensional image frame database, and the matching is successful. And when it is unsuccessful, it is effectively displayed and processed, thereby gradually enriching and perfecting the 3D image framework database, and then, with continuous use and improvement, when using the 3D image framework database, the success rate of the matching can be improved, thereby enabling the user to
- the three-dimensional image model contains more and more three-dimensional true color images, which improves the display effect.
- Embodiment 3 is a diagrammatic representation of Embodiment 3
- Step 400 Acquire image data of a user collected by an endoscope; wherein the image data includes at least image data captured by an imaging unit in the endoscope, and the type of the image data is a depth image; The image data of the collected user indicates that the stitching process has not been performed.
- Step 410 Perform splicing on the image data, and/or the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data.
- the method further includes: determining that the number of the image data is greater than a preset threshold.
- the accumulated image data is accumulated, and when the number of image data reaches a certain number, the splicing and subsequent processing are started, which can improve the processing efficiency.
- Step 420 Match the image data block including the image data in the spliced image data block with the image of the block in the saved three-dimensional image frame database, where the three-dimensional image frame database stores three-dimensional image frames.
- the method further includes:
- Step 430 If it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the spliced image data block is not obtained in the three-dimensional image frame database The image data block of the mapping relationship of the block is displayed on all or part of the spliced image data block according to the second preset display strategy.
- the display of the three-dimensional image can still be performed, and the acquired image data is spliced to obtain a plurality of mutually independent image data blocks. These three-dimensional image data blocks are displayed to meet the needs of the user's oral three-dimensional image viewing.
- the displaying, by the second preset display strategy, all or part of the spliced image data block is displayed, specifically:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block.
- Step 440 If it is determined that the preset condition is met, part or all of the spliced image data block is uploaded to the cloud server according to a preset rule, so that the cloud server is configured according to the spliced image data block. Part or all, the saved three-dimensional image frame database is updated.
- the image data block of the mapping relationship with the block in the three-dimensional image frame database has not been obtained for the endoscope since the endoscopic operation is started.
- the three-dimensional image given to the user is the spliced image data block, and only the image data block is uploaded to the cloud server, so that although the mapping relationship is not obtained, it cannot be reconstructed on the user's three-dimensional image contour, but the splicing can still be displayed.
- each image data block can be used as an initial region, so there can be multiple initial regions, and the user can freely scan any position in the oral cavity, Continuous scanning from these multiple initial regions, or scanning from other locations to form a new initial region, without continuous scanning from the unique initial region in the oral cavity, convenient for users to use, support user self-service three-dimensional true color capture Mode, improve the user experience.
- the cloud server can update the three-dimensional image frame database according to the image data blocks, for example, can be sent to the dental professional to identify and determine the image data blocks.
- Establishing a labeling system for the image data blocks wherein the labeling system may include partitioning of blocks, numbering information of the blocks, image data of the blocks, location information of the images of the blocks, and the like, wherein the location information of the images of the blocks includes : The spatial positional relationship between each block.
- the three-dimensional image framework database can be continuously updated, so that with the continuous use of the actual operation, the data included in the three-dimensional image framework database is more comprehensive and comprehensive, and after operation for a period of time, the image data block and the three-dimensional image framework database are improved.
- the image of the block is matched to the success rate, so that the user can obtain the three-dimensional image model and display the user's three-dimensional image model, which can greatly improve the user's viewing requirements and experience for the three-dimensional oral image, and the user can view the image well.
- the specific location of the oral image, the interaction effect is better.
- Embodiment 4 is a diagrammatic representation of Embodiment 4:
- Step 500 Receive an update request sent by the terminal, where the update request includes at least part or all of the image data block of the spliced image data block.
- the spliced image data block is obtained after the splicing process is performed by the terminal according to the image data of the user collected by the endoscope, and the spliced image data blocks are not obtained from the block in the three-dimensional image frame database. Mapping relationship.
- Step 510 Update the saved three-dimensional image framework database according to the update request.
- the fourth embodiment of the present invention is mainly based on the above-mentioned third embodiment of the present invention.
- the cloud server only receives the image data block, and can continuously update and improve the three-dimensional image frame database according to the image data block.
- the cloud server only receives the image data block, and can continuously update and improve the three-dimensional image frame database according to the image data block.
- the image processing method of the embodiment of the present invention is continuously updated and improved after being operated for a long time. Therefore, when matching, there may be no If the matching is unsuccessful, the image data block that has not obtained the mapping relationship will not be obtained, and the spliced image data block can obtain the mapping relationship with the block in the three-dimensional image frame database. At this time, the user's three-dimensional image is obtained.
- the model and the recorded number information of the block that has been determined to obtain the mapping relationship with the spliced image data block are uploaded to the cloud server, so that the cloud server obtains and spliced the image data according to the user's three-dimensional image model and the record.
- the cloud server can continuously update the three-dimensional image frame database according to the user's three-dimensional image model and the recorded number information of the block that has been determined to obtain the mapping relationship with the spliced image data block, which can further improve and enrich the three-dimensional image frame. database.
- Embodiment 5 is a diagrammatic representation of Embodiment 5:
- the processing loop is performed once for each image data received, and the image data block is not processed to obtain the mapping relationship with the block in the three-dimensional image framework database. And after the processing, the mapping relationship between the image data block and the block in the three-dimensional image frame database has not been described since the beginning of the endoscopic operation, wherein one processing cycle indicates that the stitching, matching, and display processing are performed once. .
- Step A) receiving the 100th image data P(100) of the user collected by the endoscope; wherein the image data includes three-dimensional true color image data captured by the camera unit in the endoscope, and the image The type of data is a depth image.
- the image data P (100), the image data block Q' (2), and the image data block Q' (5) are spliced to form a new image data block Q' (Temp).
- Step C It is judged whether or not the endoscope has been placed in the user's mouth.
- Step D) According to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the image data block of P(100) is included in the stitched image data block, that is, the image data block Q' ( Temp) matches the image of the block in the saved 3D image frame database.
- the splicing is displayed according to the second preset display strategy.
- the specific method is as follows:
- the first three image data blocks Q'(1), Q'(2), Q'(3) are taken from the sorted image data blocks. ). And displaying the three-dimensional true color image of the first three image data blocks taken out in the three display areas.
- the oral endoscopic diagram there are 3 display areas in the current display interface.
- the area 1 shows a three-dimensional true color image of Q'(1), for example, the upper left lip side block and the corresponding partial gums, the upper left lip side block and the corresponding partial gums, and the upper left upper lip side block and Corresponding part of the gums.
- Region 2 shows a three-dimensional true color image of Q'(2), for example, the right partial block on the upper right side of the lip and the corresponding gum.
- Region 3 shows a three-dimensional true color image of Q'(3), such as the lower left 4 lingual partial block and the corresponding gum.
- one processing cycle is performed every time one image data is received, and one image data block is processed in the current processing to obtain a mapping with a block in the three-dimensional image framework database. Relationship, and after adding this processing, a total of one image data block obtained from the beginning of the endoscopic operation is obtained as an example of a mapping relationship with a block in the three-dimensional image frame database, wherein one processing cycle indicates that one stitching and matching are performed. And display processing.
- Step A) receiving the 120th image data P (120) of the user collected by the endoscope; wherein the image data includes three-dimensional true color image data captured by the camera unit in the endoscope, and the image The type of data is a depth image.
- the image data P (120) can be spliced with the first image data block Q'(1) stored in the system, and can be spliced with the third image data block Q'(3) stored in the system.
- the image data P (120), the image data block Q'(1), and the image data block Q'(3) are spliced to form a new image data block Q' (Temp).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) According to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the image data block of P(120) is included in the stitched image data block, that is, the aforementioned new image data block.
- Q'(Temp) matches the image of the block in the saved 3D image frame database.
- the number information of each block in which the matching operation of the three-dimensional image frame database obtains the mapping relationship with the aforementioned new image data block Q′ (Temp) is recorded for example, the block that has been determined to obtain the mapping relationship includes: In the upper left 1 tooth distal middle block (No. 07.032.001), the upper left 1 tooth lip side block (No. 07.037.001), and the upper right 1 tooth lip side block (No. 07.042.001), the record has been determined to obtain the mapping.
- the number information of the block of the relationship is No. 07.032.001, 07.037.001, and 07.042.001.
- the system is denoted as Q°(1), and is recorded in the image data block sequence ⁇ Q°(1) ⁇ which has obtained the mapping relationship with the block in the three-dimensional image frame database. .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3 The three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area.
- the content displayed in the area 1 is: the updated three-dimensional image model of the user, for example,
- the user's three-dimensional image model is a three-dimensional image with the user's contour, and the right side of the left upper lip side block, the upper left 2 tooth lip side block and the corresponding part of the gum, the upper left 1 tooth lip side block and the corresponding part of the gums , the upper right side of the labial lip block and the corresponding gingiva, the upper right part of the left side of the lip side block and the corresponding gingiva.
- the content displayed in the area 2 is a three-dimensional true color image of the image data block Q'(1).
- the content displayed in the area 3 is a three-dimensional true color image of the image data block Q'(2).
- one processing cycle is performed every time one image data is received, and one image data block is processed in the current processing to obtain a mapping with a block in the three-dimensional image framework database. Relationship, and after adding this processing, the mapping relationship between the two image data blocks obtained in the three-dimensional image frame database since the start of the endoscopic operation is taken as an example, wherein one processing cycle indicates that one stitching and matching are performed. And display processing.
- Step A) receiving the 139th image data P (139) of the user collected by the endoscope; wherein the image data includes three-dimensional true color image data captured by the camera unit in the endoscope, and the image The type of data is a depth image;
- Step B) respectively determining the image data block sequence ⁇ Q°(1) ⁇ and the unobtained three-dimensional image data P(139) and the image data block saved in the system that have obtained the mapping relationship with the block in the three-dimensional image frame database.
- Each of the image data block sequences ⁇ Q'(1), ..., Q'(m), ..., Q'(18), M 18 ⁇ in the mapping relationship of the blocks in the image frame database Q° (1) Whether Q'(1), ..., Q'(m), ..., Q'(18) can be spliced with each other.
- the image data P (139) can be spliced with the image data block Q'(2) held in the system, and can be spliced with the image data block Q'(4) held in the system.
- the image data P (139), the image data block Q' (2), and the image data block Q' (4) are spliced to form a new image data block Q' (Temp).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) According to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the image data block of P(139) is included in the stitched image data block, that is, the aforementioned new image data block.
- Q'(Temp) matches the image of the block in the saved 3D image frame database.
- the number information of each of the blocks in the three-dimensional image frame database that has been determined to obtain the mapping relationship with the aforementioned new image data block Q' (Temp) is recorded, for example, this time with the image data block Q' (Temp)
- the block of the mapping relationship is the upper right 5 teeth buccal block (No. 07.062.001).
- the number information of the block that obtains the mapping relationship is 07.032.001, 07.037.001. , 07.042.001
- the block numbers of all the blocks that have been determined to obtain the mapping relationship with the spliced image data blocks are: 07.032.001, 07.037.001, 07.042 .001, 07.062.001.
- the system is recorded as Q°(2) and recorded in the image data block sequence that has obtained the mapping relationship with the block in the three-dimensional image frame database ⁇ Q°(1), Q 2) ⁇ .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3 The three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area.
- the content displayed in the area 1 is: the updated three-dimensional image model of the user, for example,
- the user's three-dimensional image model is a three-dimensional image contour with the user, and a three-dimensional true color image of the Q° (1) and Q° (2) image data blocks, for example, including the right side surface of the upper left lip side block, The upper left 2 teeth lip side block and the corresponding part of the gums, the upper left 1 tooth lip side block and the corresponding part of the gums, the upper right 1 tooth lip side block and the corresponding gingiva, the upper right 2 teeth lip side block left part of the surface and the corresponding gums, The right upper part of the 4 teeth cheek side block right side surface, the upper right 5 teeth buccal side block and the corresponding gums, the upper right 6 teeth cheek side block part of the surface and the corresponding gums.
- the content displayed in the area 2 is a three-dimensional true color image of the image data block Q'(1).
- the content displayed in the area 3 is a three-dimensional true color image of the image data
- one processing cycle is performed every time one image data is received, and one image data block is processed this time to obtain a mapping with a block in the three-dimensional image framework database.
- the previous two image data blocks that have obtained the mapping relationship with the blocks in the three-dimensional image frame database are exemplified, wherein one processing cycle indicates that the splicing, matching, and display processing is performed once.
- Step A) receiving the 168th image data P (168) of the user collected by the endoscope; wherein the image data includes three-dimensional true color image data captured by the camera unit in the endoscope, and the image The type of data is a depth image.
- Step B) respectively determining image data P (168) and image data blocks ⁇ Q°(1), Q°(2) ⁇ which have been determined to obtain a mapping relationship in the sequence of image data blocks stored in the system, and images not obtained with a mapping relationship
- the image data P (168) can be spliced with the image data block Q°(1) stored in the system, and can be spliced with the image data block Q°(2) stored in the system.
- the image data P (168), the image data block Q° (1), and the image data block Q° (2) are spliced to form a new image data block Q' (Temp), at which point the previous two have been obtained.
- the image data blocks that are mapped to the blocks in the three-dimensional image frame database are fused together by P(168) to form a new image data block.
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) according to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the new image data block Q′ (Temp) and the image of the block in the saved three-dimensional image frame database are performed. match.
- the block number of each block in which the mapping relationship with the aforementioned new image data block Q' (Temp) has been determined in the three-dimensional image frame database is recorded for example, the block that has been determined to obtain the mapping relationship includes: the upper left 2 teeth Lip side block (No. 07.032.001), upper left 1 tooth labial block (No. 07.037.001), upper right 1 tooth labial block (No. 07.042.001), upper right 2 tooth labial block (No. 07.042. 001), the upper right 5 teeth buccal block (No. 07.062.001), the recorded number information of the blocks that have been determined to obtain the mapping relationship are: 07.032.001, 07.037.001, 07.042.001, 07.047.001, 07.062 .001.
- the system is re-recorded as Q°(1), and the image data block sequence that has obtained the mapping relationship with the block in the three-dimensional image frame database ⁇ Q°(1) is recorded. ⁇ .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3 The three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area.
- the content displayed in the area 1 is: the updated three-dimensional image model of the user, where the The user's three-dimensional image model is a three-dimensional image contour with the user, and a three-dimensional true color image of the Q° (1) image data block, for example, including the left side 3 lip side block right side curved surface, the upper left 2 tooth lip side block and Corresponding part of the gums, upper left 1 tooth labial block and corresponding part of the gums, upper right 1 tooth labial block and corresponding gums, upper right 2 teeth lip side block and corresponding gums, upper right 2 teeth far middle adjacent block, upper right 2
- the three-dimensional true color surface between the distal middle block of the tooth and the adjacent medium block of the right upper 4 teeth, the right upper 5 teeth near the middle block, the upper right 5 teeth buccal block and the corresponding gingiva, the upper right 6 teeth cheek The partial surface of the side block and the corresponding gums.
- the content displayed in the area 2 is a three-dimensional true color image of the image data block Q'(1).
- the processing loop is performed once every time two image data are received, and the image data block is not processed in the current processing to obtain the mapping relationship with the block in the three-dimensional image framework database. And after the processing, the mapping relationship between the image data block and the block in the three-dimensional image frame database has not been described since the beginning of the endoscopic operation, wherein one processing cycle indicates that the stitching, matching, and display processing are performed once. .
- Step A) receiving the 100th and 101th image data P(100), P(101) of the user collected by the endoscope; wherein the image data includes the three-dimensional true color captured by the camera unit in the endoscope Image data, and the type of the image data is a depth image.
- Step B) respectively determining image data P(100), P(101) with each other, and with the image data block sequence ⁇ Q'(1), ..., Q'(m), ..., Q' held in the system. (18) Whether each of the image data blocks Q'(1), ..., Q'(m), ..., Q'(18) in M 18 ⁇ can be spliced with each other.
- image data P(100) is determined to be spliced with the second image data block Q'(2) stored in the system, and can be spliced with the fifth image data block Q'(5) held in the system.
- Image data P (101) cannot be spliced with any other image data block.
- Image data P (100), the image data block Q' (2), and the image data block Q' (5) are spliced to form a new image data block, which is denoted as Q' (Temp 1).
- Image data P (101) also constitutes a separate new block of image data, denoted Q' (Temp 2).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) according to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the new image data block Q' (Temp 1), Q' (Temp 2) and the saved three-dimensional image The images of the blocks in the framework database are matched.
- the first three image data blocks Q'(1), Q'(2), and Q'(3) are taken from the sorted image data blocks. And displaying the three-dimensional true color image of the first three image data blocks taken out in the three display areas.
- the oral endoscopic diagram there are 3 display areas in the current display interface.
- the content displayed in the area 1 is a three-dimensional true color image of Q'(1), for example, the right part of the right upper part of the lip side block, the upper right side of the upper right side of the adjacent block, the upper right side of the tooth, the upper right 2 teeth The upper half of the adjacent middle block, the upper half of the right upper 2 teeth lip and the corresponding gingival surface.
- the content displayed in the area 2 is a three-dimensional true color image of Q'(2), for example, a partial curved surface of the right upper 6 tooth side block.
- the content displayed in the area 3 is a three-dimensional true color image of Q'(3), for example, a partial curved surface of the left lower 3 tooth lingual block.
- one processing cycle is performed every time two image data are received, and one image data block is obtained in the current processing to obtain a mapping with a block in the three-dimensional image framework database. Relationship, and after adding this processing, a total of one image data block obtained from the beginning of the endoscopic operation is obtained as an example of a mapping relationship with a block in the three-dimensional image frame database, wherein one processing cycle indicates that one stitching and matching are performed. And display processing.
- Step A) receiving the 120th and 121th image data P(120), P(121) of the user collected by the endoscope; wherein the image data includes the three-dimensional true color captured by the camera unit in the endoscope Image data, and the type of the image data is a depth image.
- image data P (120) can be spliced with the first image data block Q'(1) stored in the system, and can be spliced with the fourth image data block Q'(4) stored in the system.
- Image data P (121) cannot be spliced with any other image data block.
- image data P (120), the image data block Q'(1), and the image data block Q'(4) are spliced to form a new image data block, which is denoted as Q' (Temp 1).
- Image data P (101) also constitutes a separate new block of image data, denoted Q' (Temp 2).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) according to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the image data block of the spliced image data block includes P (120), P (121) image data blocks, that is, The aforementioned new image data blocks Q' (Temp 1), Q' (Temp 2) are matched with the image features of the blocks in the saved three-dimensional image frame database.
- the new image data block Q' (Temp 1) obtains a mapping relationship with the block in the three-dimensional image frame database
- the new image data block Q' (Temp 2) cannot obtain the three-dimensional image.
- the mapping relationship of the blocks in the framework database Since the beginning of the endoscopic operation, the aforementioned new image data block Q' (Temp 1) is currently the only one that obtains the mapping relationship with the block in the three-dimensional image frame database.
- the block number of each block in which the matching operation of the three-dimensional image frame database obtains the mapping relationship with the aforementioned new image data block Q' (Temp 1) is recorded, for example, the block in which the mapping relationship has been determined is determined.
- the block in which the mapping relationship has been determined is determined.
- the recorded block containing the block of the mapping relationship with the spliced image data block has been determined since the start of the endoscopic operation. No. 07.045.001, 07.046.001, 07.047.001, 07.050.001.
- the system is denoted as Q°(1), and is recorded in the image data block sequence that has obtained the mapping relationship with the block in the three-dimensional image frame database ⁇ Q°(1) ⁇ .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3
- the three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area.
- the content displayed in the area 1 is: the updated user's content.
- a three-dimensional image model, the user's three-dimensional image model is a three-dimensional image contour with the user, and a three-dimensional true color curved surface image of Q° (1), for example, the right upper part 1 tooth lip side block right side surface curved surface, upper right 1 tooth The far middle adjacent block, the upper right 2 teeth near the middle block, the upper right 2 tooth lip side block and the corresponding gingival surface, the upper right 2 teeth far middle adjacent block, the upper right 3 tooth lip side block left part curved surface .
- the content displayed in the area 2 is: a three-dimensional true color image of the image data block Q'(1), for example, a partial curved surface of the lower right 6 tooth lingual block.
- the content displayed in the area 3 is: a three-dimensional true color image of the image data block Q'(2), for example, a partial curved surface of the left lower 4 tooth lingual block.
- one processing cycle is performed every time two image data is received, and one image data block is obtained in the current processing and the block in the three-dimensional image frame database is obtained.
- the mapping relationship and the addition of the two image data blocks to obtain the mapping relationship between the two image data blocks in the three-dimensional image frame database since the start of the endoscopic operation are described as an example, wherein one processing cycle indicates that the splicing is performed once. Match and display processing.
- Step A) receiving the 138th, 139th image data P(138), P(139) of the user collected by the endoscope; wherein the image data includes the three-dimensional true color captured by the camera unit in the endoscope Image data, and the type of the image data is a depth image.
- Step B) respectively determining image data blocks P(138), P(139) and their image data blocks in the sequence of image data blocks stored in the system that have been determined to obtain a mapping relationship with the blocks in the three-dimensional image frame database.
- ⁇ Q°(1) ⁇ and image data blocks ⁇ Q'(1), . . . , Q'(m), . . . , Q'(18), M not obtained with the mapping relationship with the blocks in the three-dimensional image frame database.
- Whether each of the image data blocks Q°(1), Q'(1), ..., Q'(m), ..., Q'(18) in 18 ⁇ can be spliced with each other.
- image data P (139) can be spliced with the image data block Q'(3) held in the system, and can be spliced with the image data block Q'(5) held in the system.
- Image data P (138) cannot be spliced with any other image data block.
- the stitched image data P (138) constitutes a separate new image data block, denoted as Q' (Temp1).
- Image data P (139), image data block Q' (3) and image data block Q' (5) are spliced to form a new image data block, denoted as Q' (Temp 2).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) According to a preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the image data block of the spliced image data block includes P (138) or P (139) image data blocks, That is, the aforementioned new image data blocks Q' (Temp 1) and Q' (Temp 2) are respectively matched with the images of the blocks in the saved three-dimensional image frame database.
- the aforementioned new image data block Q' (Temp 1) cannot obtain the mapping relationship with the block in the three-dimensional image frame database, and the aforementioned new image data block Q' (Temp 2) is obtained and the three-dimensional image is obtained.
- the system processes the image data block as follows:
- the block number of each of the blocks in the three-dimensional image frame database that has been determined to obtain the mapping relationship with the aforementioned new image data block Q' is recorded, for example, the block that has been determined to obtain the mapping relationship includes: The upper right 5 teeth near the middle block (No. 07.061.001), the upper right 5 teeth buccal block (No. 07.062.001).
- the number information of the block that has been recorded to obtain the mapping relationship is 07.045.001, 07.046.001, 07.047.001, and 07.050.001, and the number information of the block of the matching record is added.
- All the currently recorded number information of each block that has been determined to obtain a mapping relationship with the spliced image data block since the start of the endoscopic operation is No. 07.045.001, 07.046.001, 07.047.001, 07.050 .001, 07.061.001, 07.062.001.
- the system is recorded as Q° (2), and is recorded in the image data block sequence that has obtained the mapping relationship with the block in the three-dimensional image frame database ⁇ Q°(1) , Q°(2) ⁇ .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3
- the three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area.
- the content displayed in the area 1 is: the updated user's content.
- a three-dimensional image model the user's three-dimensional image model is a three-dimensional image contour with the user, and three-dimensional true color images of Q° (1) and Q° (2) image data blocks, such as Q° (1) three-dimensional true color
- the surface image is the upper right part of the right lip part of the upper lip, the right upper 1 tooth distal middle block, the upper right 2 tooth proximal middle block, the upper right 2 tooth lip side block and the corresponding gingival surface, upper right 2
- the distal part of the tooth is in the adjacent block, and the left part of the left upper part of the tooth is located on the left side of the tooth;
- the 3D true color surface image of Q°(2) is the right part of the right upper 4 teeth buccal block, and the upper right 5 teeth are near. Adjacent block, right upper 5 teeth buccal block and corresponding gums.
- the content displayed in the area 2 is: a three-dimensional true color image of the image data block Q'(1), for example, a partial curved surface of the left lower 5 tooth buccal block.
- the content displayed in the area 3 is: a three-dimensional true color image of the image data block Q'(2), for example, a partial curved surface of the lower left 4 tooth occlusion surface block.
- the processing loop is performed once every time two image data are received, and the mapping relationship with the block in the three-dimensional image framework database is not obtained in the current processing, but Since the beginning of the sneak operation, the mapping relationship between the two image data blocks and the blocks in the three-dimensional image frame database has been obtained before the current processing.
- the one-time processing cycle indicates that the splicing, matching, and display processing are performed once.
- Step A) receiving the 152th, 153th image data P(152), P(153) of the user collected by the endoscope; wherein the image data includes the three-dimensional true color captured by the camera unit in the endoscope Image data, and the type of the image data is a depth image.
- the image data P (152) can be spliced with the image data block Q'(3) held in the system.
- the image data P (153) cannot be spliced with any other image data block, for example, a partial curved surface of the lower left 4 tooth occlusion surface block.
- Image data P (152) and the image data block Q' (3) are spliced to form a new image data block, which is denoted as Q' (Temp 1).
- Image data P (153) constitutes a separate new block of image data, denoted Q' (Temp 2).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) According to the saved three-dimensional image frame database, according to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the spliced image data block includes P(152), P(153).
- the image data block that is, the aforementioned new image data block Q' (Temp 1), Q' (Temp 2) is matched with the image of the block in the saved three-dimensional image frame database.
- the image data block sequence ⁇ Q°(1), Q°(2) ⁇ which has obtained the mapping relationship with the block in the three-dimensional image frame database is unchanged.
- the three-dimensional image model of the currently saved user is unchanged; the number information of each block that has been determined to obtain the mapping relationship with the spliced image data block has not changed since the start of the current endoscopic operation.
- the first preset display strategy all or part of the currently saved user's three-dimensional image model and the spliced image data block except for other image data blocks that have been determined to obtain the mapping relationship are displayed, specifically:
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3
- the three-dimensional true color image of the first two image data blocks taken out and the three-dimensional image model of the currently saved user are respectively displayed in the display area, specifically:
- the content displayed in area 1 is: the currently saved user's three-dimensional image model, the user's three-dimensional image model is the three-dimensional image contour with the user, and the three-dimensional true color of the Q° (1) and Q° (2) image data blocks.
- the image for example, the 3D true color surface image of Q°(1) is, the upper right part of the right lip part of the right side of the tooth, the right upper 1 tooth far middle adjacent block, the upper right 2 tooth near the middle side block, the upper right 2 tooth lip side block and corresponding gingival surface, right upper 2 teeth far middle adjacent block, upper right 3 teeth lip side block left part of the surface;
- Q ° (2) three-dimensional true color surface image, upper right 4 teeth cheek The right side part of the side block, the upper right 5 teeth near the middle block, the upper right 5 teeth buccal block and the corresponding gums.
- the content displayed in the area 2 is: a three-dimensional true color image of the image data block Q'(1), for example, a partial curved surface of the lower left tooth occlusal surface block. Note: Since this image data block is spliced with image data P (152), its area will be larger than the area of the image data block displayed before splicing.
- the content displayed in the area 3 is: a three-dimensional true color image of the image data block Q'(2), for example, a partial curved surface of the left lower 4 tooth buccal block.
- one processing cycle is performed every time two image data are received, and one image data block is processed in the current processing to obtain a block in the three-dimensional image frame database.
- the previous two image data blocks that have obtained the mapping relationship with the blocks in the three-dimensional image frame database are merged as an example, wherein one processing cycle represents performing a stitching, matching, and display processing. .
- Step A) receiving the 168th, 169th image data P(168), P(169) of the user collected by the endoscope; wherein the image data includes the three-dimensional true color captured by the camera unit in the endoscope Image data, and the type of the image data is a depth image.
- the image data P (168) can be spliced with the image data block Q°(1) stored in the system, and can be spliced with the image data block Q°(2) stored in the system.
- Image data P (169) cannot be spliced with any other image data block.
- Image data P (168), the image data block Q° (1), and the image data block Q° (2) are spliced to form a new image data block, which is denoted as Q' (Temp 1).
- Image data P (169) constitutes a separate new block of image data, denoted Q' (Temp 2).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) according to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the image data block of the spliced image data block includes P (168), P (169) image data blocks, that is, the foregoing
- the new image data blocks Q' (Temp 1), Q' (Temp 2) are matched to the images of the blocks in the saved three-dimensional image frame database.
- the new image data block Q' (Temp 1) obtains a mapping relationship with the block in the three-dimensional image frame database
- the new image data block Q' (Temp 2) cannot obtain the three-dimensional image.
- the processing method of the image data block is as follows:
- the block number of each block in which the mapping relationship with the aforementioned new image data block Q' (Temp 1) is determined is recorded in the three-dimensional image frame database, for example, the mapping is determined to be obtained.
- the blocks of the relationship include: the upper right 1 adjacent tooth adjacent block (No. 07.041.001), the upper right 1 tooth lip side block (No. 07.042.001), and the upper right 1 tooth far middle adjacent block (No. 07.045.001) ), the upper right 2 teeth near the middle block (No. 07.046.001), the upper right 2 tooth lip side block (No. 07.047.001), the upper right 2 teeth far middle adjacent block (No. 07.050.001), the upper right 5 Near the middle face of the tooth (No.
- the number information of the recorded blocks that have been determined to obtain the mapping relationship are: 07.041.001, 07.042.001, 07.045.001, 07.046.001, 07.047.001, 07.050.001, 07.061 .001, 07.062.001.
- the system is denoted as Q°(1), and is recorded in the image data block sequence that has obtained the mapping relationship with the block in the three-dimensional image frame database ⁇ Q°(1) ⁇ .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3
- the three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area, specifically:
- the content displayed in the area 1 is: the updated three-dimensional image model of the user, the three-dimensional image model of the user is a three-dimensional image contour with the user, and the three-dimensional true color of the Q° (1) image data block. image.
- the three-dimensional true color image of the Q° (1) image data block is the upper right part of the upper right side of the tooth lip side block, the right upper 1 tooth far middle adjacent block, the upper right 2 tooth near middle adjacent block, the upper right 2 tooth lip side block and corresponding gingival surface, right upper 2 teeth far middle adjacent block; right upper 2 teeth far middle adjacent block and upper right 5 teeth near the middle adjacent block between the three-dimensional true color surface; upper right 5
- the proximal part of the tooth is adjacent to the block, the upper right 5 teeth cheek side block and the corresponding gum.
- the content displayed in the area 2 is: a three-dimensional true color image of the image data block Q'(1), for example, a partial curved surface of the left lower 5 tooth buccal block.
- the content displayed in the area 3 is: a three-dimensional true color image of the image data block Q'(2), for example, a partial curved surface of the lower left 4 tooth occlusion surface block.
- Embodiment 14 is a diagrammatic representation of Embodiment 14:
- one processing cycle is performed every time six image data is received, and one image data block is obtained in the current processing and the block in the three-dimensional image frame database is obtained.
- the mapping relationship and the addition of the two image data blocks to obtain the mapping relationship between the two image data blocks in the three-dimensional image frame database since the start of the endoscopic operation are described as an example, wherein one processing cycle indicates that the splicing is performed once. Match and display processing.
- Step A) receiving the first image data P (138), P (139), P (140), P (141), P of the 138, 139, 140, 141, 142, and 143 of the user collected by the endoscope. (142), P (143);
- Step B) respectively determining image data P (138), P (139), P (140), P (141), P (142), P (143) with each other, and the image data block saved in the system
- the image data block ⁇ Q°(1) ⁇ which has obtained the mapping relationship with the block in the three-dimensional image frame database and the image data block ⁇ Q' (1) which does not obtain the mapping relationship with the block in the three-dimensional image frame database have been determined in the sequence.
- the image data P(138), P(140), P(142) can be spliced, the image data P(139), P(141), P(143) can be spliced; the image data P(139), P The image data block formed after the splicing of (141) and P (143) can be spliced with the image data block Q'(3) stored in the system, and can be spliced with the image data block Q'(5) stored in the system; image data The image data block formed after the splicing of P(138), P(140), and P(142) cannot be spliced with any other image data block.
- the image data blocks formed by splicing the image data P (138), P (140), and P (142) constitute an independent new image data block, which is denoted as Q' (Temp 1).
- the image data P (139), P (141), P (143), image data block Q' (3), and image data block Q' (5) are spliced to form a new image data block, which is denoted as Q' ( Temp2).
- Step C) determines whether the endoscope has been placed in the user's mouth, and if so, proceeds to step D).
- Step D) according to the preset image pattern recognition algorithm, based on the image feature information of the block in the three-dimensional image frame database, the new image data block Q' (Temp 1), Q' (Temp 2) and the three-dimensional image The images of the blocks in the framework database are matched.
- the new image data block Q' (Temp 1) cannot obtain the mapping relationship with the block in the three-dimensional image frame database
- the new image data block Q' (Temp 2) is obtained and the three-dimensional image frame is obtained.
- the block number of each of the blocks in the three-dimensional image frame database that has been determined to obtain the mapping relationship with the aforementioned new image data block Q' is recorded, for example, the block that has been determined to obtain the mapping relationship includes: The upper right 5 teeth near the middle block (No. 07.061.001), the upper right 5 teeth buccal block (No. 07.062.001).
- the number information of the block that has been recorded to obtain the mapping relationship is 07.045.001, 07.046.001, 07.047.001, and 07.050.001, and the block recorded after the current processing is added.
- the number information, the number of information of each block that has been determined to obtain the mapping relationship with the spliced image data block since the current endoscope is turned on is currently numbered 07.045.001, 07.046.001, 07.047.001 , 07.050.001, 07.061.001, 07.062.001.
- the system is recorded as Q° (2), and is recorded in the image data block sequence that has obtained the mapping relationship with the block in the three-dimensional image frame database ⁇ Q°(1) , Q°(2) ⁇ .
- the first two image data blocks Q'(1), Q'(2) are taken from the sorted image data blocks, and in the 3
- the three-dimensional true color image of the first two image data blocks and the updated three-dimensional image model of the user are respectively displayed in the display area, specifically:
- the content displayed in area 1 is: the updated user's three-dimensional image model, the user's three-dimensional image block module is a three-dimensional image contour with the user, and the three-dimensional true of the Q° (1) and Q° (2) image data blocks.
- Color image For example, the 3D true color surface image of Q°(1) is the right part of the right upper part of the lip side block, the right upper 1 tooth far middle adjacent block, the upper right 2 tooth near middle part block, the upper right 2 teeth.
- the content displayed in the area 2 is: a three-dimensional true color image of the image data block Q'(1), for example, a partial curved surface of the left lower 5 tooth buccal block.
- the content displayed in the area 3 is: a three-dimensional true color image of the image data block Q'(2), for example, a partial curved surface of the lower left 4 tooth occlusion surface block.
- the image processing apparatus specifically includes:
- the receiving unit 10 is configured to acquire image data of a user collected by the endoscope; wherein the image data includes at least image data captured by an imaging unit in the endoscope, and the type of the image data is a depth image.
- the image data of the collected user indicates that the stitching process is not performed;
- the processing unit 11 is configured to perform splicing with the image data, and/or the image data and the saved image data block to obtain a spliced image data block, wherein the image data block includes at least one image data; And respectively matching the image data block including the image data in the stitched image data block with the image of the block in the saved three-dimensional image frame database, and if determining to obtain at least one image data block including the image data And a mapping relationship between the blocks in the three-dimensional image frame database, and determining, according to the mapping relationship, a position of the at least one block corresponding to the image data block including the image data in the saved three-dimensional image contour of the user Reconstructing the at least one image data block including the image data to a corresponding determined position in the three-dimensional image contour of the user, to obtain reconstructed three-dimensional image data, according to the reconstructed 3D image data, updating a currently saved three-dimensional image model of the user; wherein the number of the three-dimensional image frames
- the inventory stores data of an image of
- the display unit 12 is configured to, according to the first preset display strategy, all or part of the updated three-dimensional image model of the user and the other image data blocks of the stitched image data block that have been determined to obtain a mapping relationship.
- the sending unit 13 is configured to: if it is determined that the preset condition is met, the updated three-dimensional image model of the user and the recorded number information of the block that has been determined to obtain the mapping relationship, and according to a preset rule, And part of the image data block except the image data block that has been determined to obtain the mapping relationship with the block in the three-dimensional image frame database, and uploaded to the cloud server, so that the cloud server is updated according to the update
- the saved three-dimensional image frame database is updated by part or all of the image data blocks outside the data block.
- the location information of the image of the block includes: a spatial positional relationship between each block;
- An image of each of the three-dimensional image contours is a three-dimensional curved shape based on an image of the block in the three-dimensional image frame database or the three-dimensional image model of the user, including a preset single color and a single texture image .
- the processing unit 11 is further configured to:
- the updated three-dimensional image model of the user and all or part of the other image data blocks in the stitched image data block that have been determined to obtain a mapping relationship are displayed.
- the display unit 12 is specifically configured to:
- Arranging, according to the area of the image data block or the image data block, the number of the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship are sorted from large to small;
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data blocks, and the updated three-dimensional image model of the user is sequentially displayed in the d display regions. And the first d-1 image data blocks taken out;
- n is less than d-1
- the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as that of n+1, and the update is displayed in the display area in the adjusted current display interface.
- the processing unit 11 is further configured to:
- the number information of the block that has determined to obtain the mapping relationship is recorded according to the mapping relationship.
- the processing unit 11 is further configured to: if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and at least one of the spliced image data blocks The image data block having a mapping relationship with the block in the three-dimensional image frame database has been obtained, and the display unit 12 is further configured to:
- the currently saved three-dimensional image model of the user and all or part of the other image data blocks in the stitched image data block that have been determined to obtain a mapping relationship are displayed.
- the display unit 12 is specifically configured to:
- Arranging, according to the area of the image data block or the image data block, the number of the image data, the other image data blocks in the stitched image data block except the other image data blocks that have been determined to obtain the mapping relationship are sorted from large to small;
- n is not smaller than d-1
- the first d-1 image data blocks are taken out from the sorted image data block, and the currently saved three-dimensional image model of the user is sequentially displayed in the d display regions. And the first d-1 image data blocks taken out;
- n is less than d-1, the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as that of n+1, and the current display areas in the adjusted current display interface are sequentially displayed.
- the processing unit 11 is further configured to: if it is determined that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the spliced image data block is not obtained.
- the image data block of the mapping relationship of the blocks in the three-dimensional image framework database is further used by the display unit 12 to:
- all or part of the spliced image data block is displayed according to the second preset display strategy, and the display unit 12 is specifically configured to:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block;
- the processing unit 11 is further configured to:
- the receiving unit 10 is further configured to: after the displaying unit 12 according to the first preset display policy, determine the updated three-dimensional image model of the user and the spliced image data block. After all or part of the other image data blocks of the mapping relationship are displayed, the image data of the user acquired by the acquisition of the endoscope is returned.
- the image processing apparatus specifically includes:
- the receiving unit 20 is configured to acquire image data of a user collected by the endoscope; wherein the image data includes at least image data captured by an imaging unit in the endoscope, and the type of the image data is a depth image.
- the image data of the collected user indicates that the stitching process is not performed;
- the processing unit 21 is configured to perform splicing with the image data, and/or the image data and the saved image data block to obtain a spliced image data block, where the image data block includes at least one image data; Matching, respectively, the image data block including the image data in the spliced image data block and the image of the block in the saved three-dimensional image frame database; wherein the three-dimensional image frame database stores the image of the three-dimensional image frame
- the display unit 22 is configured to: if the processing unit determines that the mapping relationship between the image data block including the image data and the block in the three-dimensional image frame database is not obtained, and the spliced image data block is not obtained and And displaying, in the image data block of the mapping relationship of the blocks in the three-dimensional image frame database, displaying all or part of the spliced image data block according to the second preset display strategy;
- the sending unit 23 is configured to upload part or all of the spliced image data block to the cloud server according to a preset rule, if the predetermined condition is met, so that the cloud server is configured according to the spliced image. Part or all of the data block, the saved three-dimensional image frame database is updated.
- all or part of the spliced image data block is displayed according to the second preset display strategy, and the display unit 22 is specifically configured to:
- the spliced image data blocks are sorted from large to small according to the area of the image data block or the number of image data blocks included in the image data block;
- the first d image data blocks are taken out from the sorted image data blocks, and respectively In the d display areas, the first d image data blocks are sequentially displayed, and if a is smaller than d, the number of display areas in the current display interface is adjusted, so that the number of display areas is the same as a, and Displaying the a plurality of image data blocks in sequence in the display area in the adjusted current display interface;
- the receiving unit 20 is further configured to: after the displaying unit 22 displays all or part of the spliced image data block according to the second preset display policy, returning to performing execution of acquiring the user collected by the endoscope; Image data.
- the processing unit 21 is further configured to:
- the image processing apparatus in the fifteenth embodiment and the sixteenth embodiment, wherein the receiving unit, the processing unit, the display unit, and the sending unit are all integrated in one user terminal, For example, they are all integrated in the mobile phone. Of course, they can also be separated.
- the receiving unit, the processing unit and the transmitting unit can be integrated into the handle of the endoscope, and the display unit is integrated in the mobile phone;
- the receiving unit and a part of the functions of the processing unit are integrated into the handle of the endoscope, and other functions, the display unit and the sending unit of the processing unit are integrated in the mobile phone, and in actual implementation, the present invention is applicable.
- the endoscope sends the image data to the mobile phone, performs three-dimensional surface image stitching and processing on the mobile phone, and then displays the processed three-dimensional true color image on the mobile phone, which is performed locally, without connecting to the cloud through the Internet.
- it can also be implemented in the built-in engine of the mobile phone chip.
- the chip with the built-in ANN engine can also be implemented by setting the chip of the built-in ANN engine.
- the image processing method in the embodiment of the present invention may include the following operations when the actual project is implemented:
- Step 1) Acquire image data collected by the endoscope and obtain the image data block after splicing.
- Step 2) Based on the existing oral image database and the image data block in step 1), construct a human oral endoscopic panoramic label system, that is, a three-dimensional image framework database, wherein the label system of the three-dimensional image framework database includes at least the block number information. , relative position information, image information, and the like.
- Step 3) Based on the labeling system of step 2), the ANN is trained by a training method such as deep learning and/or reinforcement learning to obtain the latest version of the execution ANN, wherein the ANN represents a multi-layer neural network with a specific architecture and weights. .
- a training method such as deep learning and/or reinforcement learning to obtain the latest version of the execution ANN, wherein the ANN represents a multi-layer neural network with a specific architecture and weights.
- Step 4) Put the latest version of the execution ANN into the chip to undertake image matching processing functions, such as AI chips such as Cambrian or Kirin 970NPU.
- AI chips such as Cambrian or Kirin 970NPU.
- Step 5) splicing, matching, reconstructing and the like of the image data collected after the endoscope is placed in the oral cavity, thereby realizing the user's self-service multi-initial point endoscopic image three-dimensional true color display, that is, the user's mouth can be obtained.
- the three-dimensional image in turn, can perform an artificial intelligence check on the user's current oral health based on the three-dimensional image of the user's mouth.
- Step 6) For image data that exceeds the existing label system, one is to splicing the image data, obtaining the spliced image data block to meet the user's endoscopic needs, and the second is uploading to the cloud server and distributing it to the dental expert. To update the labeling system of the 3D image framework database. Subsequently, the whole system obtained the latest version of the oral endoscopic panoramic label system, that is, the updated three-dimensional image framework database. After that, repeat step 2).
- embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROA, optical storage, etc.) including computer usable program code.
- computer-usable storage media including but not limited to disk storage, CD-ROA, optical storage, etc.
- the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
- the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
- These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
- the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Optics & Photonics (AREA)
- Epidemiology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Processing Or Creating Images (AREA)
- Endoscopes (AREA)
- Image Analysis (AREA)
Abstract
本发明涉及图像处理领域,尤其涉及一种图像处理方法及装置,获取图像数据,进行拼接,获得拼接后的图像数据块;将拼接后的图像数据块中包含所述图像数据的图像数据块与三维图像框架数据库中区块的图像进行匹配,若获得至少一个映射关系,更新用户的三维图像模型,根据第一预设展示策略进行展示;若拼接后的图像数据块都没有获得映射关系,根据第二预设展示策略进行展示;若满足预设条件,将用户的三维图像模型、记录的已确定获得映射关系的区块的编号信息和/或未获得映射关系的图像数据块上传到云端服务器,使云端服务器更新三维图像框架数据库,无需对口腔连续有序扫描,就可以得到三维图像,并不断完善三维图像框架数据库,提高匹配成功率。
Description
本申请要求在2017年11月01日提交中国专利局、申请号为201711058265.7、发明名称为“一种图像处理方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本发明涉及图像处理技术领域,尤其涉及一种图像处理方法及装置。
传统的口腔内窥器是用于牙科光学取模的一种设备,在使用时需要把取模用的光学枪头在用户上牙列、下牙列有序移动,不支持扫描枪头在口腔内随意漫游,专业性很强,用户互动效果差,需要专业人士才能操作。
目前,也出现了一种用户自助式的口腔内窥器,用户手持这种设备,将摄像头部分放入口腔,转动摄像头对口腔进行内窥。但内窥图像每次只能看到一个很小的局部,虽然也能看到牙齿的画面,但难以确认这是哪一颗牙的画面,也不知道当前所看到的细节处于口腔的哪个具体位置。并且,缺乏三维图像信息,无法对接全口牙列取模,无法实时生成当下的全口牙列数字化模型,因而也无法支撑牙套式洁齿器3D打印等牙科应用。
发明内容
本发明实施例提供一种图像处理方法及装置,以解决现有技术中口腔图像呈现的用户互动效果差、难以支持用户自助式的三维真彩取模的问题。
本发明实施例提供的具体技术方案如下:
一种图像处理方法,包括:
步骤A:获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;
步骤B:针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;
步骤C:分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;
步骤D:若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数 据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型,以及,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示;其中,所述用户的三维图像模型的初始值为所述用户的三维图像轮廓;
步骤E:若确定满足预设条件,则将更新后的所述用户的三维图像模型和记录的已确定获得映射关系的区块的编号信息,以及根据预设的规则,将所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器根据更新后的所述用户的三维图像模型和已确定获得映射关系的区块的编号信息,以及所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新保存的所述三维图像框架数据库。
较佳的,所述区块的图像的位置信息包括:每个区块相互之间的空间位置关系;
所述三维图像轮廓中每一个区块的图像,为基于所述三维图像框架数据库或所述用户的三维图像模型中区块的图像的三维曲面形状,包括预设的单一颜色和单一纹理的图像。
较佳的,分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配之前,进一步包括:
根据拼接后的图像数据块和预设的图像模式识别算法,确定所述内窥器已放入口腔。
较佳的,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体包括:
根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;
若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示更新后的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示更新后的所述用户的三维图像模型和排序后的图像数据块。
较佳的,进一步包括:
若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,记录已确定获得映射关系的区块的编号信息。
较佳的,若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中有至少一个已获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则所述步骤D进一步包括:
根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
较佳的,根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体包括:
根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;
若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示当前保存的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示当前保存的所述用户的三维图像模型和排序后的图像数据块。
较佳的,若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则所述步骤D进一步包括:
根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
较佳的,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,具体包括:
若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a 个图像数据块。
较佳的,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,进一步包括:
确定所述图像数据的个数大于预设阈值。
较佳的,进一步包括:
执行步骤D之后,返回执行所述步骤A。
一种图像处理方法,包括:
步骤A:获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;
步骤B:针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;
步骤C:分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;
步骤D:若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示;
步骤E:若确定满足预设条件,则根据预设的规则,将所述拼接后的图像数据块的部分或全部,上传到云端服务器,以使云端服务器根据所述拼接后的图像数据块的部分或全部,更新保存的所述三维图像框架数据库。
较佳的,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,具体包括:
若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块;
所述方法进一步包括:执行步骤D之后,返回执行所述步骤A。
较佳的,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,进一步包括:
确定所述图像数据的个数大于预设阈值。
一种图像处理装置,包括:
接收单元,用于获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;
处理单元,用于针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;并分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,以及若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型;其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述用户的三维图像模型的初始值为所述用户的三维图像轮廓;
展示单元,用于根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示;
发送单元,用于若确定满足预设条件,则将更新后的所述用户的三维图像模型和记录的已确定获得映射关系的区块的编号信息,以及根据预设的规则,将所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器根据更新后的所述用户的三维图像模型和已确定获得映射关系的区块的编号信息,以及所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新保存的所述三维图像框架数据库。
较佳的,所述区块的图像的位置信息包括:每个区块相互之间的空间位置关系;
所述三维图像轮廓中每一个区块的图像,为基于所述三维图像框架数据库或所述用户的三维图像模型中区块的图像的三维曲面形状,包括预设的单一颜色和单一纹理的图像。
较佳的,分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配之前,处理单元进一步用于:
根据拼接后的图像数据块和预设的图像模式识别算法,确定所述内窥器已放入口腔。
较佳的,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,展示 单元具体用于:
根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;
若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示更新后的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示更新后的所述用户的三维图像模型和排序后的图像数据块。
较佳的,处理单元进一步用于:
若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,记录已确定获得映射关系的区块的编号信息。
较佳的,处理单元进一步用于若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中有至少一个已获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则展示单元进一步用于:
根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
较佳的,根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,展示单元具体用于:
根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;
若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示当前保存的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示当前保存的所述用户的三维图像模型和排序后的图像数据块。
较佳的,处理单元进一步用于若确定未获得包含所述图像数据的图像数据块与所述三 维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则展示单元进一步用于:
根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
较佳的,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,展示单元具体用于:
若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块。
较佳的,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,处理单元进一步用于:
确定所述图像数据的个数大于预设阈值。
较佳的,所述接收单元进一步用于在所述展示单元根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示之后,返回执行获取内窥器采集到的用户的图像数据。
一种图像处理装置,包括:
接收单元,用于获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;
处理单元,用于针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;并分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配;其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;
展示单元,用于若处理单元确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示;
发送单元,用于若确定满足预设条件,则根据预设的规则,将所述拼接后的图像数据块的部分或全部,上传到云端服务器,以使云端服务器根据所述拼接后的图像数据块的部分或全部,更新保存的所述三维图像框架数据库。
较佳的,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,展示单元具体用于:
若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块;
所述接收单元进一步用于在所述展示单元根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示之后,返回执行获取内窥器采集到的用户的图像数据;
较佳的,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,处理单元进一步用于:
确定所述图像数据的个数大于预设阈值。
本发明实施例中,获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据,并且所述图像数据的类型为深度图像;所述采集到的用户的图像数据表示未进行拼接处理的;针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述区块的图像数据包括:编号信息、图像特征信息;根据不同的匹配结果,执行相应的不同的展示方法和上传到云端服务器不同的信息,以使云端服务器更新三维图像框架数据库,这样,对获取到的用户的图像数据进行拼接,获得拼接后的图像数据块,根据建立的三维图像框架数据库,对拼接后的图像数据块进行匹配处理,若至少一个匹配成功,则根据用户的三维图像轮廓,进行重构,更新当前保存的用户的三维图像模型,进而对用户的三维图像模型和未获得映射关系的图像数据块进行展示,若均未匹配成功,则展示拼接后的图像数据块,不需要内窥器对口腔连续有序扫描,用户可以使用内窥器随意扫描口腔,只要获得口腔内表面各处图像数据而不论是否有序,并且不管是否匹配成功,都可以展示口腔的三维图像,不需要专业人士操作,很好地支持了用户自助式口腔内窥,以及不仅可以呈现用户口腔的三维图像,而且可以动态进行展示,展示效果更好,提升了用户的使用体验和互动效果,可以很好地支持用户自助式的三维真彩取模。
并且,根据不同的匹配结果,向云端服务器可能发送不同的信息,但都可以使云端服 务器根据这些信息更新三维图像框架数据库,解决了口腔的三维图像处理有条件执行区块匹配和展示的问题,可以不断更新三维图像框架数据库,从而逐步充实和完善三维图像框架数据库,流程更加完整,更能接近实际的工程实现,进而随着不断的使用和更新,在之后使用三维图像框架数据库时,可以提高匹配的成功率,从而可以使得用户的三维图像模型中包含的三维真彩图像越来越多,提高了展示效果。
图1为本发明实施例一中图像处理方法流程图;
图2为本发明实施例一中当前界面中的展示区域划分的示意图;
图3为本发明实施例二中云端服务器的更新方法流程图;
图4为本发明实施例三中图像处理方法流程图;
图5为本发明实施例四中云端服务器的更新方法流程图;
图6为本发明实施例九中三维图像展示效果示意图;
图7为本发明实施例十中三维图像展示效果示意图;
图8为本发明实施例十一中三维图像展示效果示意图;
图9为本发明实施例十三中三维图像展示效果示意图;
图10为本发明实施例十五中图像处理装置结构示意图;
图11为本发明实施例十六中图像处理装置结构示意图。
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,并不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
值得说明的是,本发明实施例中,主要针对口腔三维图像,其中,内窥器可以为口腔内窥器,当然,本发明实施例中,并不仅限于口腔三维图像,对于其它领域的三维图像同样也可以适用,下面仅以口腔为例进行说明。
实施例一:
参阅图1所示,本发明实施例一中,图像处理方法的具体流程如下:
步骤100:获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据,并且所述图像数据的类型为深度图像;所述采集到的用户的图像数据表示未进行拼接处理的。
其中,内窥器,例如为口腔内窥器,该内窥器中设置有摄像单元,用于拍摄图像,内窥器中可以设置有一个摄像单元,也可以设置有多个摄像单元,并且拍摄到的图像数据的类型为深度图像,也就是RGBD图像,为三维真彩图像,这样可以获取图像的三维信息,便于后续进行三维图像处理。
较佳的,本发明实施例中的口腔内窥器,可以在内窥头上背靠背设置有两个摄像单元,这两个摄像单元相对位置固定并预先标定、同步照明、同步摄像。这样,这两个摄像单元同步拍摄到的图像数据分别带有对应的摄像单元标识,例如摄像单元的编号信息,并且具有相同的时序戳信息,其中,时序戳信息表示从内窥器开机以来,第几次采集到图像数据,这样,可以针对每次采集到的图像数据,都设置时序戳,这样也便于区分不同时间采集到的图像数据。
这样,步骤100中获取到的图像数据为多个带有摄像单元标识和时序戳信息的未进行拼接处理的图像数据,这些图像数据可能带有相同的时序戳,也可能带有不同的时序戳,这根据内窥器中摄像单元的数目,即内窥器每次采集到的图像数据的数目,和设置的每次进行处理的图像数据的数目而定。
进一步地,执行步骤110之前,还包括:确定图像数据的个数大于预设阈值。
例如,内窥器每次采集到2个图像数据,预设阈值为6,即设置的每次进行处理的图像数据的数目为6个,则需要获取到内窥器3次拍摄采集到的用户的图像数据,才执行后续的拼接处理,例如,这6个图像数据分别为图像数据P(1)、P(2)、P(3)、P(4)、P(5)、P(6),其中,P(1)、P(2)是时序戳相同(r=1)而摄像单元编号分别为1和2的两幅图像数据;P(3)、P(4)是时序戳相同(r=2)而摄像单元编号分别为1和2的两幅图像数据;P(5)、P(6)是时序戳相同(r=3)而摄像单元编号分别为1和2的两幅图像数据。
也就是说,本发明实施例中,为了提高处理效率,并不一定是每次接收到内窥器采集到的图像数据就进行后续处理,因为内窥器一次采集到的图像数据可能比较少,并且,通常内窥器采集到的一个图像数据都是面积很小的一个曲面,因此,可以在达到一定个数后再进行处理,可以提高后续拼接、匹配等处理的准确性和效率。
步骤110:针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据。
本发明实施例中,将图像数据相互之间以及图像数据与保存的图像数据块相互之间分别进行判断和拼接,得到拼接后的图像数据块,进行保存,并根据拼接后的图像数据块更新保存的图像数据块,也就是说,本发明实施例中,拼接后的图像数据块,表示所有能够拼接和不能够拼接的拼接处理后的图像数据,其中,不仅包括能够拼接,并拼接成功、与其他图像数据拼接后组成的更大面积的图像数据,还包括经过判断后拼接不成功,依旧孤立存在的图像数据,不仅包括包含步骤100中获取到的图像数据即本次待处理的图像数据 的图像数据块,还包括没有包含步骤100中获取到的图像数据的图像数据块,图像数据经过拼接处理后变换成图像数据块,拼接后的图像数据块中可以包括一个图像数据,表示不能够拼接的单个图像数据,也可以包括多个图像数据,表示这多个图像数据能够拼接并拼接后得到的更大的图像数据。
例如,接收到第N个图像数据P(N),经过拼接处理后获得拼接后的图像数据块可以为T+M个相互独立(即相互不能拼接)的图像数据块,分别为Q°(1)、…、Q°(t)、…、Q°(T),以及Q(1)、Q(m)、…、Q(M),其中,图像数据块Q°(t)由n°(t)个图像数据拼接而成,图像数据块Q(m)由n(m)个图像数据拼接而成,并且,[n°(1)+…+n°(t)+…n°(T)]+[n(1)+n(2)+…n(m)+…n(M)]=N;n(t)≥1,n(m)≥1。
进一步地,本发明实施例中,还可以记录拼接后的图像数据块包括的图像数据的个数,这样以便于后续对图像数据块进行排序。
这样,本发明实例中,先将获取到的图像数据进行拼接,再执行后续的匹配,拼接后的图像数据块面积也相对较大,可以提高匹配效率,也可以减少图像数目,减少图像与区块的匹配次数,减少时间,也提高了执行效率。
步骤120:分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述区块的图像的数据包括:编号信息、图像特征信息。
本发明实施例中,预先建立了口腔的三维图像框架数据库和三维图像轮廓。
1)本发明实施例中,三维图像框架数据库,基于人类口腔的各种情况进行构建,该三维图像框架数据库存储了人类口腔的三维图像模型的通用框架数据,该框架数据涵盖了各种情况下人类口腔全部表面区域的图像特征信息,例如形状特征、色彩特征、纹理特征等信息。
三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像数据以及每个区块的图像的位置信息;区块的图像的位置信息包括:每个区块相互之间的空间位置关系;区块的图像数据包括:编号信息、图像特征信息。
2)本发明实施例中,三维图像轮廓存储了人类全口腔内表面各个区域(含各个区块)的三维图像的形状轮廓数据。其中,用户的三维图像轮廓至少存储了所述用户的口腔中各个区块的三维图像的形状轮廓数据。并且,本发明实施例中,在实际使用时,可以根据每个用户的实际情况,更新该用户的三维图像轮廓,随着不同用户的使用,针对每个用户,就会有属于自己口腔的三维图像轮廓。
所述三维图像轮廓中每一个区块的图像,为基于所述三维图像框架数据库或所述用户的三维图像模型中区块的图像的三维曲面形状,包括预设的单一颜色和单一纹理的图像。
执行步骤120时,具体包括:可以根据区块的图像特征信息、区块相互之间的空间位置关系和/或编号信息、或者根据图像数据块中图像数据携带的摄像单元的标识,确定的拼接后的图像数据块的相对空间位置关系,分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配。
其中,本发明实施例中,本次仅将拼接后的图像数据块中包含步骤100中图像数据的图像数据块与三维图像框架数据库中区块的图像进行匹配,其中,步骤100中的图像数据即是本次待处理的图像数据,这是因为,本次进行拼接处理之前,当前保存的图像数据块必定都是经过匹配处理后的,而本发明实施例中,若匹配成功的图像数据块都会重构于用户的三维图像轮廓中,因此当前保存的都是匹配未成功的图像数据块,若本次进行拼接处理后,是将所有的拼接后的图像数据块进行匹配,对于那些与新的图像数据不能拼接、仍是之前保存时的图像数据块,匹配也必定是不成功的,再次进行匹配只会增加匹配次数,降低处理效率,因此,本发明实施例中,对于这部分图像数据块就不需要再次进行匹配了,也节省时间,提高效率。
例如,获取内窥器采集到的用户的图像数据为P(1)和P(2),保存的图像数据块为{Q′(1)、…、Q′(m)、…、Q′(32),M=32},则针对P(1)、P(2)以及P(1)、P(2)与Q′(1)、…、Q′(m)、…、Q′(32)进行拼接,若图像数据P(1)与系统中保存的第1幅图像数据块Q′(1)能拼接、且与系统中保存的第4幅图像数据块Q′(4)能拼接,图像数据P(2)与任何其他图像数据块都无法拼接,则对图像数据P(1)、图像数据块Q′(1)、图像数据块Q′(4)进行拼接,形成新的图像数据块,记为Q′(Temp 1),图像数据P(2)也构成一个独立的新的图像数据块,记为Q′(Temp 2),则在匹配时,仅将Q′(Temp 1)和Q′(Temp 2)与三维图像框架数据库中区块的图像进行匹配。
进一步地,执行步骤120之前,还包括:
根据拼接后的图像数据块和预设的图像模式识别算法,确定所述内窥器已放入口腔。
这种方法,通过图像模式识别来判断内窥器是否已放入口腔,不需要用户通过开关手动开启是否开始进行口腔内窥,在实现时内窥器上就无需设置该开关,减少了硬件的设置,同时也提高了用户的使用体验。
当然,也可以采用其它方法判断内窥器是否已放入口腔,例如,在内窥器上设置开关,由用户手动控制,用户把内窥器放入口腔后,手动闭合开关,则系统认为内窥器已经放入口腔。
这样,本发明实施例中,主要是针对口腔的三维图像处理,因此,根据图像模式识别算法,若确定图像数据块不是口腔图像,则可以不用继续执行后续的匹配过程,避免无效的操作,若确定图像数据块为口腔图像,则可以判定内窥器已放入口腔,可以继续执行后续的匹配操作,
步骤130:若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型,以及,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示;其中,所述用户的三维图像模型的初始值为所述用户的三维图像轮廓。
执行步骤130时,具体包括:
首先,若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型。
其中,所述用户的三维图像模型的初始值为所述用户的三维图像轮廓。
具体包括:1)根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置。
具体为:至少根据三维图像框架数据库中区块的图像特征信息,确定所述至少一个包含所述图像数据的图像数据块对应于用户的三维图像框架数据库中的区块,以及,根据所述至少一个包含所述图像数据的图像数据块对应于用户的三维图像框架数据库中的区块,确定所述至少一个包含所述图像数据的图像数据块对应于用户的三维图像轮廓中的位置。
当然,还可以结合区块相互之间的空间位置关系和/或编号信息、图像数据块相互之间的空间位置关系等方式,来确定在用户的三维图像轮廓中的位置,本发明实施例中并不进行限制。
2)将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据。
具体为:根据三维图像框架数据库中区块的边界特征信息,从所述至少一个包含所述图像数据的图像数据块中提取出属于对应的区块的三维曲面图像;其中,所述图像特征信息中至少包括区块的边界特征信息;将提取出的三维曲面图像替换用户的三维图像轮廓中对应的确定的位置上的图像,获得重构后的三维图像数据。
3)根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型。
具体为:将重构后的三维图像数据,替换掉当前保存的用户的三维图像模型中对应的 确定的位置上的图像。
这样,每次得到重构后的三维图像数据后,都可以不断地替换用户的三维图像模型中相应位置上的图像,实现动态更新用户的三维图像模型的效果。
进一步地,还可以更新用户的三维图像轮廓,具体为:
根据更新后的用户的三维图像模型,获取更新后的用户的三维图像模型对应的三维图像轮廓,并根据更新后的所述用户的三维图像模型对应的三维图像轮廓,更新保存的用户的三维图像轮廓。
这样,可以针对不同的用户,会有相应的符合该用户的口腔三维图像轮廓,在之后再进行扫描口腔时,可以更容易看到该用户实际的口腔图像信息,并且,不断更新用户的三维图像模型和该用户的三维图像轮廓,也可以提高匹配的效率和准确性。
值得说明的是,本发明实施例中,一开始建立的三维图像框架数据库,可能并不是很全面,无法包含所有情况下的口腔图像,这样在将图像数据块与三维图像框架数据库中区块的图像进行匹配时,很有可能出现匹配不成功的情况,因此,本发明实施例中,主要针对实际中三维图像框架数据库的建立问题,解决了口腔三维图像处理时有条件执行区块匹配和展示的问题,解决如何不断更新和完善三维图像框架数据库,以及解决在匹配成功和不成功时如何进行展示和处理的问题,从而能够逐步充实、完善三维图像框架数据库,流程更加完整,更能接近实际的工程实现。
这样,在执行步骤130时,若确定获得至少一个包含所述图像数据的图像数据块与三维图像框架数据库中区块的映射关系,就可以更新用户的三维图像模型,实现后续三维图像模型的展示。
进一步地,还包括:
若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,记录已确定获得映射关系的区块的编号信息。
本发明实施例中,记录已确定获得映射关系的区块的编号信息,是为了用于之后更新三维图像框架数据库,可以对三维图像框架数据库中区块的编号或位置、或区块的图像数据、区块的图像特征信息,也包括增加一个或一组区块等区块级别的操作等信息进行更新。
然后,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
具体包括:1)根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序。
本发明实施例中,图像数据块是由一个或多个图像数据组成,因此,通常图像数据块包括图像数据的个数越多,图像数据块的面积越大,当然也可能包括个数较少的图像数据块的面积反而更大,这对于之后的排序和展示没有影响,本发明实施例中,可以根据包括 的图像数据的个数进行排序,也可以根据其面积进行排序,对此并不进行限制,可以择一方式进行排序。
其中,由于本发明实施例中,采集到的图像数据为深度图像,并携带有摄像单元标识,因此可以根据摄像单元的内标定参数和深度图像数据,计算获取图像数据块的面积,具体地计算图像数据块的面积的方法,可以基于现有技术中的方法,本发明实施例中,并不进行限制,这里就不再详述了。
例如,拼接后的图像数据块共有m个,其中,拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数为n,将拼接后的图像数据块中除已确定获得映射关系的其它图像数据块,按照包括图像数据的个数,从大到小排序后为Q′(1)、…、Q′(n)。
本发明实施例中,进行排序的目的是为了后续进行展示时,能够优先展示那些面积比较大或者包括图像数据个数比较多的图像数据块,因为面积越大或包括的图像数据个数越多,体现的图像信息就越多,便于用户查看和理解展示的图像。
2)确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小。
这里判断n与d-1的关系,是因为这时已经有图像数据块与区块匹配成功了,能够对用户的三维图像模型进行更新,在展示时需要预留一个区域来展示用户的三维图像模型,因此,对于未匹配成功的图像数据块的展示区域有d-1个。
3)若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示更新后的所述用户的三维图像模型和所述取出的前d-1个图像数据块。
例如,n为5,d为3,判断5>(3-1),则需要从排序后Q′(1)、…、Q′(n)中取出前2个即Q′(1)和Q′(2),在展示时,参阅图2所示,为当前界面中的展示区域的划分,分为区域1、区域2和区域3,则在第一个展示区域,即区域1中展示更新后的用户的三维图像模型,在第二个展示区域,即区域2中展示Q′(1),在第三个展示区域,即区域3中展示Q′(2)。
4)若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示更新后的所述用户的三维图像模型和排序后的图像数据块。
例如,n为2,d为4,判断2<(4-1),则说明当前展示区域的个数比较多,会有没有展示图像的空白区域,为进一步提升用户观看体验和提升展示效果,因此,需要减少展示区域的个数,将d减少为3,当前展示界面中变为有3个展示区域,则在展示时,在第一个展示区域中展示更新后的用户的三维图像模型,在第二个展示区域中展示Q′(1),在第三个展示区域中展示Q′(2)。
本发明实施例中,在执行步骤130时,除了上述若确定获得至少一个包含所述图像数 据的图像数据块与所述三维图像框架数据库中区块的映射关系的情况之外,还包括以下几种情况:
第一种情况:若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中有至少一个已获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
本发明实施例中,从口腔内窥器开机,使用口腔内窥器开始进行内窥操作以来,会不断地获取到口腔内窥器采集到的图像数据,每当图像数据个数达到预设阈值,就开始执行一次拼接和匹配操作,直到内窥操作结束,这里的第一种情况,是针对本次进行拼接和匹配处理后,包含本次处理的图像数据的图像数据块没有获得与三维图像框架数据库中区块的映射关系,但是从口腔内窥器开机,使用口腔内窥器开始进行内窥操作以来,在本次进行拼接和匹配处理之前的某次处理中,已有获得与三维图像框架数据库中区块的映射关系的情况,这时,虽然本次不会对用户的三维图像模型进行更新,但是当前仍有保存的用户的三维图像模型,在展示时仍需要展示当前保存的用户的三维图像模型和本次拼接后的图像数据块。
具体包括:1)根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序。
2)确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小。
3)若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示当前保存的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
4)若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示当前保存的所述用户的三维图像模型和排序后的图像数据块。
这里的第一种情况中的第一预设展示策略与步骤130中第一预设展示策略相同,只是展示的内容不同而已,因此,这里就不再详细进行介绍了。
第二种情况:若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
这里的第二种情况,是针对从口腔内窥器开机,使用口腔内窥器开始进行内窥操作以 来,一直未获得与三维图像框架数据库中区块的映射关系的情况。
本发明实施例中,拼接后的图像数据块是多个相互独立的图像数据块。这是因为,本发明实施例中,内窥器设置有两个或以上的摄像单元,同步采集图像,因此本发明实施例中的图像处理方法支持用户自助式完成口腔内窥操作,无需用户有序、连续扫描。
具体包括:1)若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序。
2)确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块。
例如,将拼接后的图像数据块从大到小排序后,分别为Q′(1)、…、Q′(a),若a为6,d为4,判断6>4,则从Q′(1)、…、Q′(a)中取出前4个图像数据块,分别为Q′(1)、Q′(2)、Q′(3)和Q′(4),在展示时,在第一个展示区域中展示Q′(1),在第二个展示区域中展示Q′(2),在第三个展示区域中展示Q′(3),在第四个展示区域中展示Q′(4)。
又例如,若a为3,d为4,判断3<4,则需要减少展示区域的个数,将d减少为3,当前展示界面中变为有3个展示区域,则在展示时,在第一个展示区域中展示Q′(1),在第二个展示区域中展示Q′(2),在第三个展示区域中展示Q′(3)。
进一步地,当再次获取到内窥器采集到的用户的图像数据,执行步骤130之后,返回执行上述步骤100。
这样,本发明实施例中,从内窥器开机,使用内窥器开始进行内窥操作以来,会不断地获取到内窥器采集到的图像数据,通过不断的拼接和匹配,更新当前展示的图像数据块和/或用户的三维图像模型,提升展示效果,用户可以看到用户的三维图像模型,对于匹配未成功的图像数据块,用户可以看到展示的拼接后的图像数据块。
步骤140:若确定满足预设条件,则将更新后的所述用户的三维图像模型和记录的已确定获得映射关系的区块的编号信息,以及根据预设的规则,将所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器根据更新后的所述用户的三维图像模型和已确定获得映射关系的区块的编号信息,以及所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新保存的所述三维图像框架数据库。
其中,预设条件,可以为以下几种:确定内窥器关机、或用户的三维图像模型中已重构的区域的面积大于预设阈值、或拼接后的图像数据块的面积达到预设阈值等,本发明实 施例中,并不进行限制,这样,本发明实施例中,不是每次更新完用户的三维图像模型和得到新的已确定获得映射关系的区块的编号信息就进行上传,而是达到一定条件时再统一进行上传,不仅可以减少上传次数,减少服务器压力,也可以提高云端服务器对三维图像框架数据库进行更新时的效率。
其中,上传到云端服务器的已确定获得映射关系的区块的编号信息,不仅包括本次进行拼接和匹配处理后记录的获得映射关系的区块的编号信息,还包括从内窥操作开始,每次进行拼接和匹配处理记录的所有已确定获得映射关系的区块的编号信息。
其中,上述预设的规则,例如为拼接后的图像数据块包括图像数据的个数不小于预设数目,例如5个,或者,拼接后的图像数据块的面积不小于预设面积,例如100mm
2,本发明实施例中并不进行限制,可以根据实际情况进行设置。
也就是说,本发明实例中,上传到云端服务器的信息可以包括以下几部分:1)用户的三维图像模型。2)记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息。3)拼接后的图像数据块中除已确定获得与三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块。这样,云端服务器可以根据这些信息,不断更新三维图像框架数据库,具体为更新其中区块的编号信息、增加区块、增加不同情况下人类口腔的三维图像模型的通用框架数据等,逐渐更新和完善三维图像框架数据库,形成一个包括强大的完整的口腔三维图像框架数据库,具有完整的标签体系,提高了之后使用该三维图像框架数据库匹配的成功率。
值得说明的是,本发明实施例中,在实际实现过程中在不同阶段上传到云端服务器的信息可能是不同的,例如可以将实际实现过程分为三个阶段:
阶段一:在初期,由于三维图像框架数据库还很不完善,可能拼接后的图像数据块都不能与三维图像框架数据库中区块的图像匹配成功,也就不能获得与拼接后的图像数据块的映射关系的区块,因此,可能无法获得上述第一部分和二部分信息即1)用户的三维图像模型和2)记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,这时,上传到云端服务器的信息可能只有上述第三部分信息即3)拼接后的图像数据块中除已确定获得与保存的三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,进而这时云端服务器也只能根据拼接后的图像数据块中除已确定获得与保存的三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新三维图像框架数据库。
阶段二:随着不断使用和更新三维图像框架数据库,会提高匹配成功率,至少一个图像数据块能与三维图像框架数据库中区块的图像匹配成功,进而就会更新用户的三维图像模块,并可以获得已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,当然这时也很有可能会有不能匹配成功的图像数据块,因此,这时上传到云端服务器的信息会 包括:1)用户的三维图像模型。2)记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息。3)拼接后的图像数据块中除已确定获得与保存的三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,进而这时云端服务器可以根据这三部分信息,更新三维图像框架数据库。
阶段三:随着本发明实施例的图像处理方法运营很长时间后,可以达到成熟阶段,这时三维图像框架数据库已经比较完善了,能够包括所有口腔情况的图像,并三维图像框架数据库的标签体系也很完善,其中,标签体系至少包括区块的编号信息、相对位置信息、图像信息等,因此,在匹配时,成功率会极大提升,可能不会有匹配不成功的情况,因此,这时可能不会获得3)拼接后的图像数据块中除已确定获得与保存的三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器的信息也就只有1)用户的三维图像模型和2)记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,进而这时云端服务器根据用户的三维图像模型和记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,更新三维图像框架数据库。
本发明实施例中,针对获取到的内窥器采集到的用户的图像数据,以及该图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,然后与保存的三维图像框架数据库中区块的图像进行匹配,根据不同的匹配结果和相应的不同的展示策略,展示用户的三维图像模型和拼接后的图像数据块,最后若确定满足预设条件时,将用户的三维图像模型、记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,以及拼接后的图像数据块中除已确定获得与三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器可以不断更新三维图像框架数据库,解决了口腔的三维图像处理时有条件执行区块匹配和展示的问题,可以不断更新和完善三维图像框架数据库,以及在匹配成功和不成功时有效地进行展示和处理,从而逐步充实和完善三维图像框架数据库,提高匹配成功率,进而随着三维图像框架数据库的不断完善,在后期进行匹配时,基本都可以匹配成功,可以不断将匹配成功的图像数据块重构于用户的三维图像轮廓中对应的确定的位置上,并更新用户的三维图像模型,这样,展示的用户的三维图像模型上会包含越来越多的三维真彩图像,若没有匹配不成功的图像数据块,则在展示时就会仅展示用户的三维图像模型,而用户的三维图像模型的展示效果更好,使得用户可以更直观查看到口腔的图像的具体部位;本发明实施例中,不仅可以获得用户口腔的三维图像,使得用户可以查看口腔的具体部位,而且,由于可以获得多个相互独立的图像数据块,每一个图像数据块都可以作为一个初始区域,因此可以有多个初始区域,用户可以随意扫描口腔中的任意位置,可以从这多个初始区域开始进行连续扫描,也可以从其它位置扫描从而形成一个新的初始区域,无需在口腔中从唯一的初始区域连续扫描,便于用户使用,提升了用户的使用体验,进而动态展示扫描到的口腔的三维图像,呈 现效果更好,更加方便灵活。
实施例二:
基于上述实施例,参阅图3所示,本发明实施例二中,云端服务器的更新方法的具体流程如下:
步骤300:接收终端发送的更新请求,其中,所述更新请求中至少包括:用户的三维图像模型、记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,以及拼接后的图像数据块中除已确定获得与保存的三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块。
其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像数据以及每个区块的图像的位置信息;所述区块的图像数据包括:编号信息、图像特征信息;所述拼接后的图像数据块为所述终端根据内窥器采集到的用户的图像数据进行拼接处理后得到的。
本发明实施例中,在实际实现过程中,更新请求中可能会包含不同的信息,具体地在上述实施例一中有详细介绍,这里就不再赘述了。
步骤310:根据所述更新请求,更新保存的所述三维图像框架数据库。
本发明实施例二中,主要针对拼接后的图像数据块与三维图像框架数据库的区块既有匹配成功的,也有未匹配成功时,云端服务器接收到三种信息,即用户的三维图像模型、区块的编号信息和未获得映射关系的图像数据块的情况。
本发明实施例中,在更新三维图像框架数据库时,具体地:
1)在用户的三维图像模型中,已确定获得与拼接后的图像数据块的映射关系的各个区块之外的其他三维真彩曲面图像部分,即是超出现有三维图像框架数据库中(现有标签体系)已知区块标签的图像。
对于这部分三维真彩曲面图像,在内窥过程中,本发明实施例的处理方法是通过执行传统的图像拼接,填补了已确定与拼接后的图像数据块的映射关系的各个区块之间的空隙,由此满足了用户口腔内窥和三维建模的需求。但是,本发明实施例中,仍需要将用户的三维图像模型和记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息发送给云端服务器。
云端服务器接收到之后,需要检测以下情况:用户的三维图像模型中的已确定获得映射关系的区块之间存在尚未被识别的其他三维真彩曲面图像部分。然后,云端服务器要把存在尚未被识别的其他三维真彩曲面图像部分的、用户的三维图像模型分发给相应的牙科专家,以便相应的牙科专家对那些尚未被识别的其他三维真彩曲面图像部分进行研究分析,并更新标签体系,从而更新得到最新的三维图像框架数据库(口腔内窥全景标签体系)。
2)对于拼接后的图像数据块中除已确定获得与保存的三维图像框架数据库中区块的 映射关系的图像数据块之外的部分或全部图像数据块,云端服务器也可根据预先设定的策略,把这些图像数据块分发给相应的牙科专家,以便相应的牙科专家对这些图像数据块进行研究分析,并更新标签体系,从而更新得到最新的三维图像框架数据库(口腔内窥全景标签体系)。
这样,本发明实施例中,云端服务器可以不断更新三维图像框架数据库,解决了口腔的三维图像处理有条件执行区块匹配和展示的问题,可以不断更新和完善三维图像框架数据库,以及在匹配成功和不成功时有效地进行展示和处理,从而逐步充实和完善三维图像框架数据库,进而随着不断的使用和完善,在之后使用三维图像框架数据库时,可以提高匹配的成功率,从而可以使得用户的三维图像模型中包含的三维真彩图像越来越多,提高了展示效果。
实施例三:
基于上述实施例,参阅图4所示,本发明实施例三中,图像处理方法的具体流程如下:
步骤400:获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据,并且所述图像数据的类型为深度图像;所述采集到的用户的图像数据表示未进行拼接处理的。
步骤410:针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据。
进一步地,执行步骤410之前,还包括:确定所述图像数据的个数大于预设阈值。
这样,累计获取到的图像数据,当图像数据的个数达到一定数量后,才开始执行本次拼接和后续处理,可以提高处理效率。
步骤420:分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述区块的图像的数据包括:编号信息、图像特征信息。
进一步地,执行步骤420之前进一步包括:
根据拼接后的图像数据块和预设的图像模式识别算法,确定所述内窥器已放入口腔。
这样,针对口腔内窥的情况,根据图像模式识别,自动识别是否已放入口腔,并且当确定已放入口腔后再执行后续操作,避免无效的操作,节省处理资源,提高效率。
步骤430:若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
也就是说,进行匹配后,不仅包含本次处理的图像数据的图像数据块没有获得映射关 系,而且内窥器开机开始进行内窥操作以来,拼接后的图像数据块都没有获得映射关系,这时就无法重构更新用户的三维图像模型,对于这种情况,本发明实施例中,仍可以进行三维图像的展示,是对获取到的图像数据进行拼接,得到多个相互独立的图像数据块,展示的是这些三维的图像数据块,满足用户的口腔三维图像查看的需求。
其中,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,具体包括:
1)若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序。
2)确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块。
步骤440:若确定满足预设条件,则根据预设的规则,将所述拼接后的图像数据块的部分或全部,上传到云端服务器,以使云端服务器根据所述拼接后的图像数据块的部分或全部,更新保存的所述三维图像框架数据库。
也就是说,本发明实施例三中,主要针对内窥器本次开始进行内窥操作以来,一直未有获得与三维图像框架数据库中区块的映射关系的图像数据块的情况,这时展示给用户的三维图像为拼接后的图像数据块,并且上传到云端服务器的也只有图像数据块,这样,虽然未获得映射关系,不能重构于用户的三维图像轮廓上,但仍可以展示拼接后的图像数据块,并且,由于图像数据块是多个独立的三维图像,每一个图像数据块都可以作为一个初始区域,因此可以有多个初始区域,用户可以随意扫描口腔中的任意位置,可以从这多个初始区域开始进行连续扫描,也可以从其它位置扫描从而形成一个新的初始区域,无需在口腔中从唯一的初始区域连续扫描,便于用户使用,支持用户自助式的三维真彩取模,提升了用户的使用体验。
并且通过将这些未获得映射关系的图像数据块上传到云端服务器,云端服务器可以根据这些图像数据块,更新三维图像框架数据库,例如,可以发送给牙科专业人士,对这些图像数据块进行辨认判断,为这些图像数据块建立标签体系,其中标签体系可以包括区块的划分、区块的编号信息、区块的图像数据、区块的图像的位置信息等,其中,区块的图像的位置信息包括:每个区块相互之间的空间位置关系。进而,可以不断更新充实三维图像框架数据库,这样,随着实际运营的不断使用,三维图像框架数据库包括的数据更为丰富全面,运营一段时间后,就提高了图像数据块与三维图像框架数据库中区块的图像进行匹配的成功率,从而可以重构获得用户的三维图像模型,展示用户的三维图像模型,可以 极大提高用户对三维口腔图像的查看需求和体验,用户可以很好地查看到口腔图像的具体位置,互动效果更好。
实施例四:
基于上述实施例,参阅图5所示,本发明实施例四中,云端服务器的更新方法的具体流程如下:
步骤500:接收终端发送的更新请求,其中,所述更新请求中至少包括:拼接后的图像数据块的部分或全部图像数据块。
其中,所述拼接后的图像数据块为终端根据内窥器采集到的用户的图像数据进行拼接处理后得到的,并且所述拼接后的图像数据块均未获得与三维图像框架数据库中区块的映射关系。
步骤510:根据所述更新请求,更新保存的三维图像框架数据库。
本发明实施例四主要基于上述本发明实施例三的情况,这时云端服务器只接收到了图像数据块,进而可以根据图像数据块,不断更新和完善三维图像框架数据库,具体地可以参阅上述实施例一至三,这里就不再进行赘述了。
值得说明的是,在上述实施例一中也提到,随着本发明实施例的图像处理方法运营很长时间后,三维图像框架数据库不断更新和完善,因此,在匹配时,可能不会有匹配不成功的情况,也就不会获得未获得映射关系的图像数据块,拼接后的图像数据块都可以获得与三维图像框架数据库中区块的映射关系,这时,是将用户的三维图像模型和记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息上传到云端服务器,以使云端服务器根据用户的三维图像模型和记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,更新三维图像框架数据库。进而这时云端服务器可以根据用户的三维图像模型和记录的已确定获得与拼接后的图像数据块的映射关系的区块的编号信息,不断更新三维图像框架数据库,可以更加完善和充实三维图像框架数据库。
实施例五:
基于上述实施例中,本发明实施例五中,是以每接收到1幅图像数据就做1次处理循环,并且本次处理没有图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来也没有图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第100幅图像数据P(100);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(100)与系统中保存的图像数据块序列,例如为{Q′(1)、…、Q′(m)、…、Q′(9),M=9}中的每一个图像数据块Q′(1)、…、Q′(m)、…、 Q′(9)相互之间是否能拼接。
经判断,确定该图像数据P(100)与保存的第2幅图像数据块Q′(2)能拼接、且与保存的第5幅图像数据块Q′(5)能拼接。
因此,将图像数据P(100)、图像数据块Q′(2)、图像数据块Q′(5)进行拼接,形成新的图像数据块Q′(Temp)。
步骤C)判断内窥器是否已放入用户口腔。
若为“否”,对于保存的图像数据块Q′(1)、Q′(3)、Q′(4)、Q′(6)、Q′(7)、Q′(8)、Q′(9),以及新的图像数据块Q′(Temp),按包括的图像数据的个数或面积从大到小顺序进行排列,由此获得当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(8),M=8}并予以保存。返回执行步骤A)。
若为“是”,继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(100)的图像数据块,即图像数据块Q′(Temp)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若无法获得前述新的图像数据块Q′(Temp)与三维图像框架数据库中区块的映射关系(也即T=0),则根据第二预设展示策略,展示拼接后的图像数据块,具体方法如下:
对于当前保存的图像数据块Q′(1)、Q′(3)、Q′(4)、Q′(6)、Q′(7)、Q′(8)、Q′(9),以及新的图像数据块Q′(Temp),从大到小顺序排序后获得当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(8),M=8}并予以保存。
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前3个图像数据块Q′(1)、Q′(2)、Q′(3)。并在该3个展示区域中分别展示取出的前3个图像数据块的三维真彩图像。
系统返回前述步骤A)。
例如,口腔内窥示意图:当前展示界面中有3个展示区域。
其中,区域1展示的是Q′(1)的三维真彩图像,例如为左上1唇侧区块及相应部分牙龈、左上2唇侧区块及相应部分牙龈、部分左上3唇侧区块及相应部分牙龈。区域2展示的是Q′(2)的三维真彩图像,例如为右上1唇侧的右侧部分区块及相应牙龈。区域3展示的是Q′(3)的三维真彩图像,例如为左下4舌侧部分区块及相应牙龈。
实施例六:
基于上述实施例中,本发明实施例六中,是以每接收到1幅图像数据就做1次处理循环,并且本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来共有1个图像数据块获得与三维图像框架数据库中区块 的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第120幅图像数据P(120);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(120)与系统中保存的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(16),M=16}中的每一个图像数据块Q′(1)、…、Q′(m)、…、Q′(16)相互之间是否能拼接。
经判断,确定该图像数据P(120)与系统中保存的第1幅图像数据块Q′(1)能拼接、且与系统中保存的第3幅图像数据块Q′(3)能拼接。
因此,将图像数据P(120)、图像数据块Q′(1)、图像数据块Q′(3)进行拼接,形成新的图像数据块Q′(Temp)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(120)的图像数据块,即前述新的图像数据块Q′(Temp)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp)获得与三维图像框架数据库中区块的映射关系,并且,本次内窥操作开始以来,前述新的图像数据块Q′(Temp)是目前唯一的1个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=1)。则系统对图像数据块的处理方法如下:
根据该映射关系,确定前述新的图像数据块Q′(Temp)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp)重构于用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对三维图像框架数据库中本次匹配操作获得与前述新的图像数据块Q′(Temp)的映射关系的各个区块的编号信息进行记录,例如,已确定获得映射关系的区块包括:左上1牙齿远中邻面区块(编号07.032.001)、左上1牙齿唇侧区块(编号07.037.001)、右上1牙齿唇侧区块(编号07.042.001),则记录已确定获得映射关系的区块的编号信息为,编号07.032.001、07.037.001、07.042.001。
对于前述新的图像数据块Q′(Temp),系统记为Q°(1),并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定获得映射关系的其它图像数据块Q′(2)、Q′(4)、…….、Q′(16),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′ (14),M=14}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:区域1展示的内容为:更新后的用户的三维图像模型,例如该用户的三维图像模型为,带有用户的三维图像轮廓,以及左上3唇侧区块右侧部分曲面、左上2牙齿唇侧区块及相应部分牙龈、左上1牙齿唇侧区块及相应部分牙龈、右上1牙齿唇侧区块及相应牙龈、右上2牙齿唇侧区块左侧部分曲面及相应牙龈。区域2展示的内容为:图像数据块Q′(1)的三维真彩图像。区域3展示的内容为:图像数据块Q′(2)的三维真彩图像。
系统返回前述步骤A)。
实施例七:
基于上述实施例中,本发明实施例七中,是以每接收到1幅图像数据就做1次处理循环,并且本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来共有2个图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第139幅图像数据P(139);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像;
步骤B)分别判断图像数据P(139)与系统中保存的图像数据块中已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)}和未获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(18),M=18}中的每一个图像数据块Q°(1)、Q′(1)、…、Q′(m)、…、Q′(18)相互之间是否能拼接。
经判断,确定该图像数据P(139)与系统中保存的图像数据块Q′(2)能拼接、且与系统中保存的像数据块Q′(4)能拼接。
因此,将图像数据P(139)、图像数据块Q′(2)、图像数据块Q′(4)进行拼接,形成新的图像数据块Q′(Temp)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(139)的图像数据块,即前述新的图像数据块Q′(Temp)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp)获得与三维图像框架数据库中区 块的映射关系,并且,本次内窥操作开始以来,前述新的图像数据块Q′(Temp)是目前第2个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=2)。则系统对图像数据块的处理方法如下:
根据该映射关系,确定前述新的图像数据块Q′(Temp)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp)重构于用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对三维图像框架数据库中已确定获得与前述新的图像数据块Q′(Temp)的映射关系的各个区块的编号信息进行记录,例如,本次与图像数据块Q′(Temp)获得映射关系的区块为右上5牙齿颊侧区块(编号07.062.001)。
这时,例如本次拼接匹配处理之前,即第1个获得与三维图像框架数据库中区块的映射关系的图像数据块时,获得映射关系的区块的编号信息为07.032.001、07.037.001、07.042.001,则本次内窥操作开始以来,记录的所有已确定获得与拼接后的图像数据块的映射关系的各个区块的区块编号为:编号07.032.001、07.037.001、07.042.001、07.062.001。
对于前述新的图像数据块Q′(Temp),系统记为Q°(2)并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)、Q°(2)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定获得映射关系的其它图像数据块Q′(1)、Q′(3)、Q′(5)、…….、Q′(18),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(16),M=16}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:区域1展示的内容为:更新后的用户的三维图像模型,例如,该用户的三维图像模型为,带有用户的三维图像轮廓,以及Q°(1)和Q°(2)图像数据块的三维真彩图像,例如包括左上3唇侧区块右侧部分曲面、左上2牙齿唇侧区块及相应部分牙龈、左上1牙齿唇侧区块及相应部分牙龈、右上1牙齿唇侧区块及相应牙龈、右上2牙齿唇侧区块左侧部分曲面及相应牙龈,右上4牙齿颊侧区块右侧部分曲面、右上5牙齿颊侧区块及相应牙龈、右上6牙齿颊侧区块部分曲面及相应牙龈。区域2展示的内容为:图像数据块Q′(1)的三维真彩图像。区域3展示的内容为:图像数据块Q′(2)的三维真彩图像。
系统返回前述步骤A)。
实施例八:
基于上述实施例中,本发明实施例八中,是以每接收到1幅图像数据就做1次处理循环,并本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系且本次处理后,使得之前两个已获得与三维图像框架数据库中区块的映射关系的图像数据块融合为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第168幅图像数据P(168);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(168)与系统中保存的图像数据块序列中已确定获得映射关系的图像数据块{Q°(1)、Q°(2)}和未获得映射关系的图像数据块{Q′(1)、…、Q′(m)、…、Q′(12),M=12}中的每一个图像数据块Q°(1)、Q°(2)、Q′(1)、…、Q′(m)、…、Q′(12)相互之间是否能拼接。
经判断,确定该图像数据P(168)与系统中保存的图像数据块Q°(1)能拼接、且与系统中保存的像数据块Q°(2)能拼接。
因此,将图像数据P(168)、图像数据块Q°(1)、图像数据块Q°(2)进行拼接,形成新的图像数据块Q′(Temp),此时说明之前两个已获得与三维图像框架数据库中区块的映射关系的图像数据块通过P(168)融合在了一起,形成了一个新的图像数据块。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将前述新的图像数据块Q′(Temp)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp)获得与三维图像框架数据库中区块的映射关系,并且,本次内窥操作开始以来,前述新的图像数据块Q′(Temp)是目前唯一的1个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=1)。则系统对图像数据块的处理方法如下:
根据该映射关系,确定前述新的图像数据块Q′(Temp)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp)重构于用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对三维图像框架数据库中已确定与前述新的图像数据块Q′(Temp)的映射关系的各个区块的区块编号予以记录,例如已确定获得映射关系的区块包括:左上2牙齿唇侧区块(编号07.032.001)、左上1牙齿唇侧区块(编号07.037.001)、右上1牙齿唇侧区块(编号07.042.001)、右上2牙齿唇侧区块(编号07.042.001)、右上5牙齿颊侧区块(编号 07.062.001),则记录的已确定获得映射关系的区块的编号信息为:07.032.001、07.037.001、07.042.001、07.047.001、07.062.001。
对于前述新的图像数据块Q′(Temp),系统重新记为Q°(1),并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定获得映射关系的其它图像数据块Q′(1)、Q′(2)、Q′(3)、…….、Q′(12),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(12),M=12}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:区域1展示的内容为:更新后的用户的三维图像模型,其中该用户的三维图像模型为带有用户的三维图像轮廓,以及Q°(1)图像数据块的三维真彩图像,例如包括左上3唇侧区块右侧部分曲面、左上2牙齿唇侧区块及相应部分牙龈、左上1牙齿唇侧区块及相应部分牙龈、右上1牙齿唇侧区块及相应牙龈、右上2牙齿唇侧区块及相应牙龈、右上2牙齿远中邻面区块,右上2牙齿远中邻面区块与右上4牙齿近中邻面区块之间的三维真彩曲面,右上5牙齿近中邻面区块、右上5牙齿颊侧区块及相应牙龈、右上6牙齿颊侧区块部分曲面及相应牙龈。区域2展示的内容为:图像数据块Q′(1)的三维真彩图像。区域3展示的内容为:图像数据块Q′(2)的三维真彩图像。
系统返回前述步骤A)。
实施例九:
基于上述实施例中,本发明实施例九中,是以每接收到2幅图像数据就做1次处理循环,并且本次处理没有图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来也没有图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第100和101幅图像数据P(100)、P(101);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(100)、P(101)相互之间,及其与系统中保存的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(18),M=18}中的每一个图像数据块Q′(1)、…、Q′(m)、…、Q′(18)相互之间是否能拼接。
经判断,确定图像数据P(100)与系统中保存的第2幅图像数据块Q′(2)能拼接、且 与系统中保存的第5幅图像数据块Q′(5)能拼接。图像数据P(101)与任何其他图像数据块都无法拼接。
因此,将图像数据P(100)、图像数据块Q′(2)、图像数据块Q′(5)进行拼接,形成新的图像数据块,记为Q′(Temp 1)。图像数据P(101)也构成一个独立的新的图像数据块,记为Q′(Temp 2)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若无法获得前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)与三维图像框架数据库中区块的映射关系(并且,本次内窥操作以来,之前也还没有哪一个图像数据块获得与三维图像框架数据库中区块的映射关系,也即T=0),则根据第二预设展示策略,展示拼接后的图像数据块,具体方法如下:
对于当前保存的图像数据块Q′(1)、Q′(3)、Q′(4)、Q′(6)、Q′(7)、……、Q′(18),以及新的图像数据块Q′(Temp 1)、Q′(Temp 2),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(18),M=18}并予以保存。
若当前展示界面中展示区域的个数为3,即d=3,从排序后的图像数据块中取前3个图像数据块Q′(1)、Q′(2)、Q′(3),并在该3个展示区域中分别展示取出的前3个图像数据块的三维真彩图像。
例如,参阅图6所示,口腔内窥示意图:当前展示界面中有3个展示区域。
其中,区域1展示的内容为Q′(1)的三维真彩图像,例如为右上1牙齿唇侧区块右侧部分曲面、右上1牙齿远中邻面区块上半部曲面、右上2牙齿近中邻面区块上半部曲面、右上2牙齿唇侧区块上半部及相应牙龈曲面。区域2展示的内容为Q′(2)的三维真彩图像,例如为右上6牙齿腭侧区块的部分曲面。区域3展示的内容为Q′(3)的三维真彩图像,例如为左下3牙齿舌侧区块的部分曲面。
系统返回前述步骤A)。
实施例十:
基于上述实施例中,本发明实施例十中,是以每接收到2幅图像数据就做1次处理循环,并且本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来共有1个图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第120、121幅图像数据P(120)、P(121);其中, 所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(120)、P(121)相互之间,及其与系统中保存的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(32),M=32}中的每一个图像数据块Q′(1)、…、Q′(m)、…、Q′(32)相互之间是否能拼接。
经判断,确定图像数据P(120)与系统中保存的第1幅图像数据块Q′(1)能拼接、且与系统中保存的第4幅图像数据块Q′(4)能拼接。图像数据P(121)与任何其他图像数据块都无法拼接。
因此,将图像数据P(120)、图像数据块Q′(1)、图像数据块Q′(4)进行拼接,形成新的图像数据块,记为Q′(Temp 1)。图像数据P(101)也构成一个独立的新的图像数据块,记为Q′(Temp 2)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)跟据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(120)、P(121)的图像数据块,即前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)与保存的三维图像框架数据库中区块的图像特征进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp 1)获得与三维图像框架数据库中区块的映射关系,并前述新的图像数据块Q′(Temp 2)无法获得与三维图像框架数据库中区块的映射关系,本次内窥操作开始以来,前述新的图像数据块Q′(Temp 1)是目前唯一的1个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=1),则系统对图像数据块的处理方法如下:根据该映射关系,确定前述新的图像数据块Q′(Temp 1)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp 1)重构于用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对三维图像框架数据库中本次匹配操作获得与前述新的图像数据块Q′(Temp 1)的映射关系的各个区块的区块编号予以记录,例如,已确定获得映射关系的区块包括:右上1牙齿远中邻面区块(编号07.045.001)、右上2牙齿近中邻面区块(编号07.046.001)、右上2牙齿唇侧区块及相应牙龈曲面(编号07.047.001)、右上2牙齿远中邻面区块(编号07.050.001),则记录的包含本次内窥操作开始以来,已确定获得与拼接后的图像数据块的映射关系的各个区块的区块编号为,编号07.045.001、07.046.001、07.047.001、07.050.001。
对于前述新的图像数据块Q′(Temp 1),系统记为Q°(1),并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定获得映射关系的其它 图像数据块Q′(2)、Q′(3)、Q′(5)、…….、Q′(32),以及新的图像数据块Q′(Temp 2),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(31),M=31}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:参阅图7所示,区域1展示的内容为:更新后的用户的三维图像模型,该用户的三维图像模型为带有该用户的三维图像轮廓,以及Q°(1)的三维真彩曲面图像,例如为右上1牙齿唇侧区块右侧部分曲面、右上1牙齿远中邻面区块、右上2牙齿近中邻面区块、右上2牙齿唇侧区块及相应牙龈曲面、右上2牙齿远中邻面区块、右上3牙齿唇侧区块左侧部分曲面。
区域2展示的内容为:图像数据块Q′(1)的三维真彩图像,例如为右下6牙齿舌侧区块的部分曲面。
区域3展示的内容为:图像数据块Q′(2)的三维真彩图像,例如为左下4牙齿舌侧区块的部分曲面。
系统返回前述步骤A)。
实施例十一:
基于上述实施例中,本发明实施例十一中,是以每接收到2幅图像数据就做1次处理循环,并且本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来共有2个图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第138、139幅图像数据P(138)、P(139);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(138)、P(139)相互之间,及其与系统中保存的图像数据块序列中已确定获得与三维图像框架数据库中区块的映射关系的图像数据块{Q°(1)}和未获得与三维图像框架数据库中区块的映射关系的图像数据块{Q′(1)、…、Q′(m)、…、Q′(18),M=18}中的每一个图像数据块Q°(1)、Q′(1)、…、Q′(m)、…、Q′(18)相互之间是否能拼接。
经判断,确定图像数据P(139)与系统中保存的图像数据块Q′(3)能拼接、且与系统中保存的像数据块Q′(5)能拼接。图像数据P(138)与任何其他图像数据块都无法拼接。
因此,拼接处理后图像数据P(138)构成一个独立的新的图像数据块,记为Q′(Temp1)。并将图像数据P(139)、图像数据块Q′(3)和图像数据块Q′(5)进行拼接,形成新的图像数据块,记为Q′(Temp 2)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于所述三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(138)或P(139)的图像数据块,即前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)分别与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp 1)无法获得与三维图像框架数据库中区块的映射关系,并且前述新的图像数据块Q′(Temp 2)获得与三维图像框架数据库中区块的映射关系。本次内窥操作开始以来,前述新的图像数据块Q′(Temp 2)是目前第2个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=2),则系统对图像数据块的处理方法如下:
根据映射关系,确定前述新的图像数据块Q′(Temp 2)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp 2)重构于用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对三维图像框架数据库中已确定获得与前述新的图像数据块Q′(Temp 2)的映射关系的各个区块的区块编号予以记录,例如,已确定获得映射关系的区块包括:右上5牙齿近中邻面区块(编号07.061.001)、右上5牙齿颊侧区块(编号07.062.001)。
例如,本次匹配之前,已有记录的获得映射关系的区块的编号信息为07.045.001、07.046.001、07.047.001、07.050.001,则加上本次匹配记录的区块的编号信息,目前记录的所有包含本次内窥操作开始以来,已确定获得与拼接后的图像数据块的映射关系的各个区块的编号信息为,编号07.045.001、07.046.001、07.047.001、07.050.001,07.061.001、07.062.001。
对于前述新的图像数据块Q′(Temp 2),系统记为Q°(2),并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)、Q°(2)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定获得映射关系的其它图像数据块Q′(1)、Q′(2)、Q′(4)、Q′(6)、…….、Q′(18),以及新的图像数据块Q′(Temp 1),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(17),M=17}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:参阅图8所示,区域1展示的内容为:更新后的用户的三维图像模型,该用户的三维图像模型为带有用户的三维图像轮廓,以及Q°(1)和Q°(2)图像数据块的三维真彩图像,例如Q°(1)的三维真彩曲面图像为,右上1牙齿唇侧区块右侧部分曲面、右上1牙齿远中邻面区块、右上2牙齿近中邻面区块、右上2牙齿唇侧区块及相应牙龈曲面、右上2牙齿远中邻面区块、右上3牙齿唇侧区块左侧部分曲面;Q°(2)的三维真彩曲面图像为,右上4牙齿颊侧区块右侧部分曲面、右上5牙齿近中邻面区块、右上5牙齿颊侧区块及相应牙龈。
区域2展示的内容为:图像数据块Q′(1)的三维真彩图像,例如为左下5牙齿颊侧区块的部分曲面。
区域3展示的内容为:图像数据块Q′(2)的三维真彩图像,例如为左下4牙齿咬合面区块的部分曲面。
系统返回前述步骤A)。
实施例十二:
基于上述实施例中,本发明实施例十二中,是以每接收到2幅图像数据就做1次处理循环,并且本次处理没有获得与三维图像框架数据库中区块的映射关系,但是内窥操作开始以来,在本次处理之前已有2个图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第152、153幅图像数据P(152)、P(153);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(152)、P(153)相互之间、及其与系统中保存的图像数据块序列中已确定获得映射关系的图像数据块{Q°(1)、Q°(2)}和为获得映射关系的图像数据块{Q′(1)、…、Q′(m)、…、Q′(21),M=21}中的每一个图像数据块Q°(1)、Q°(2)、Q′(1)、…、Q′(m)、…、Q′(21)相互之间是否能拼接。
经判断,确定该图像数据P(152)与系统中保存的图像数据块Q′(3)能拼接。图像数据P(153)与任何其他图像数据块都无法拼接,例如为左下4牙齿咬合面区块的部分曲面。
因此,将图像数据P(152)、图像数据块Q′(3)进行拼接,形成新的图像数据块,记为Q′(Temp 1)。图像数据P(153)构成一个独立的新的图像数据块,记为Q′(Temp 2)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据保存的三维图像框架数据库,根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(152)、P(153) 的图像数据块,即前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)都无法获得与三维图像框架数据库中区块的映射关系,本次内窥操作开始以来,已获得与三维图像框架数据库中区块的映射关系的图像数据块依旧是系统中保存的图像数据块序列{Q°(1)、Q°(2)}中的图像数据块Q°(1)和Q°(2),则系统对图像数据块的处理方法如下:
已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)、Q°(2)}不变。
当前保存的用户的三维图像模型不变;其中记录的本次内窥操作开始以来、已确定获得与拼接后的图像数据块的映射关系的各个区块的编号信息不变。
对于当前保存的未获得与三维图像框架数据库中区块的映射关系的图像数据块Q′(1)、Q′(2)、Q′(4)、Q′(5)、…….、Q′(21),以及新的图像数据块Q′(Temp 2),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(21),M=21}并予以保存。
根据第一预设展示策略,对当前保存的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和当前保存的用户的三维图像模型,具体地:
区域1展示的内容为:当前保存的用户的三维图像模型,该用户的三维图像模型为带有用户的三维图像轮廓,以及Q°(1)和Q°(2)图像数据块的三维真彩图像,例如,Q°(1)的三维真彩曲面图像为,右上1牙齿唇侧区块右侧部分曲面、右上1牙齿远中邻面区块、右上2牙齿近中邻面区块、右上2牙齿唇侧区块及相应牙龈曲面、右上2牙齿远中邻面区块、右上3牙齿唇侧区块左侧部分曲面;Q°(2)的三维真彩曲面图像为,右上4牙齿颊侧区块右侧部分曲面、右上5牙齿近中邻面区块、右上5牙齿颊侧区块及相应牙龈。
区域2展示的内容为:图像数据块Q′(1)的三维真彩图像,例如为左下5牙齿咬合面区块的部分曲面。注:因为这个图像数据块拼接了图像数据P(152),所以其面积会比拼接之前展示的该图像数据块的面积要大些。
区域3展示的内容为:图像数据块Q′(2)的三维真彩图像,例如为左下4牙齿颊侧区块的部分曲面。
系统返回前述步骤A)。
实施例十三:
基于上述实施例中,本发明实施例十三中,是以每接收到2幅图像数据就做1次处理循 环,并本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系且本次处理后,使得之前两个已获得与三维图像框架数据库中区块的映射关系的图像数据块融合为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
步骤A)接收内窥器采集到的用户的第168、169幅图像数据P(168)、P(169);其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据,并且所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(168)、P(169)相互之间、及其与系统中保存的图像数据块序列中已获得映射关系的图像数据块{Q°(1)、Q°(2)}和未获得映射关系的图像数据块{Q′(1)、…、Q′(m)、…、Q′(12),M=12}中的每一个图像数据块Q°(1)、Q°(2)、Q′(1)、…、Q′(m)、…、Q′(12)相互之间是否能拼接。
经判断,确定该图像数据P(168)与系统中保存的图像数据块Q°(1)能拼接、且与系统中保存的像数据块Q°(2)能拼接。图像数据P(169)与任何其他图像数据块都无法拼接。
因此,将图像数据P(168)、图像数据块Q°(1)、图像数据块Q°(2)进行拼接,形成新的图像数据块,记为Q′(Temp 1)。图像数据P(169)构成一个独立的新的图像数据块,记为Q′(Temp 2)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将拼接后的图像数据块中包含P(168)、P(169)的图像数据块,即前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)与保存的三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp 1)获得与三维图像框架数据库中区块的映射关系,并前述新的图像数据块Q′(Temp 2)无法获得与三维图像框架数据库中区块的映射关系。前述新的图像数据块Q′(Temp 1)是内窥操作开始以来,目前唯一的1个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=1),则系统对图像数据块的处理方法如下:
根据映射关系,确定前述新的图像数据块Q′(Temp 1)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp 1)重构于用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对本次匹配操作时,三维图像框架数据库中已确定获得与前述新的图像数据块Q′(Temp 1)的映射关系的各个区块的区块编号予以记录,例如,已确定获得映射关系的区块包括:右上1牙齿近中邻面区块(编号07.041.001)、右上1牙齿唇侧区块(编号07.042.001)、右上1牙齿远中邻面区块(编号07.045.001)、右上2牙齿近中邻面区块(编号07.046.001)、 右上2牙齿唇侧区块(编号07.047.001)、右上2牙齿远中邻面区块(编号07.050.001)、右上5牙齿近中邻面区块(编号07.061.001)、右上5牙齿颊侧区块(编号07.062.001)。则本次内窥操作开始以来,记录的已确定获得映射关系的区块的编号信息为,编号07.041.001、07.042.001、07.045.001、07.046.001、07.047.001、07.050.001、07.061.001、07.062.001。
对于前述新的图像数据块Q′(Temp 1),系统记为Q°(1),并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定映射关系的其它图像数据块Q′(1)、Q′(2)、Q′(3)、…….、Q′(12),以及新的图像数据块Q′(Temp2),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(13),M=13}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:
参阅图9所示,区域1展示的内容为:更新后的用户的三维图像模型,该用户的三维图像模型为带有用户的三维图像轮廓,以及Q°(1)图像数据块的三维真彩图像。例如,Q°(1)图像数据块的三维真彩图像为,右上1牙齿唇侧区块右侧部分曲面、右上1牙齿远中邻面区块、右上2牙齿近中邻面区块、右上2牙齿唇侧区块及相应牙龈曲面、右上2牙齿远中邻面区块;右上2牙齿远中邻面区块与右上5牙齿近中邻面区块之间的三维真彩曲面;右上5牙齿近中邻面区块、右上5牙齿颊侧区块及相应牙龈。
区域2展示的内容为:图像数据块Q′(1)的三维真彩图像,例如为左下5牙齿颊侧区块的部分曲面。
区域3展示的内容为:图像数据块Q′(2)的三维真彩图像,例如为左下4牙齿咬合面区块的部分曲面。
系统返回前述步骤A)。
实施例十四:
基于上述实施例中,本发明实施例十四中,是以每接收到6幅图像数据就做1次处理循环,并且本次处理有1个图像数据块获得与三维图像框架数据库中区块的映射关系,以及加上本次处理之后,内窥操作开始以来共有2个图像数据块获得与三维图像框架数据库中区块的映射关系为例进行说明,其中,1次处理循环表示进行一次拼接、匹配和展示处理。
例如,每接收到F组时序戳相同而摄像单元编号不同的H幅图像数据,例如F=3,H=2,即F*H=3*2=6幅图像数据时,就执行后续处理步骤,也即每次处理3*2=6幅新获取到的图像 数据。
步骤A)接收内窥器采集到的用户的第138、139、140、141、142、143合计6幅图像数据P(138)、P(139)、P(140)、P(141)、P(142)、P(143);
其中,所述图像数据包括所述内窥器中摄像单元拍摄到的三维真彩图像数据并且P(138)、P(139)是时序戳相同(r=69)而摄像单元编号分别为1和2的两幅图像数据;P(140)、P(141)是时序戳相同(r=70)而摄像单元编号分别为1和2的两幅图像数据;P(142)、P(143)是时序戳相同(r=71)而摄像单元编号分别为1和2的两幅图像数据;并且,所述图像数据的类型为深度图像。
步骤B)分别判断图像数据P(138)、P(139)、P(140)、P(141)、P(142)、P(143)相互之间,及其与系统中保存的图像数据块序列中已确定获得与三维图像框架数据库中区块的映射关系的图像数据块{Q°(1)}和没有获得与三维图像框架数据库中区块的映射关系的图像数据块{Q′(1)、…、Q′(m)、…、Q′(18),M=18}中的每一个图像数据块Q°(1)、Q′(1)、…、Q′(m)、…、Q′(18)相互之间是否能拼接。
经判断,确定图像数据P(138)、P(140)、P(142)能拼接、图像数据P(139)、P(141)、P(143)能拼接;图像数据P(139)、P(141)、P(143)拼接之后形成的图像数据块与系统中保存的图像数据块Q′(3)能拼接、且与系统中保存的像数据块Q′(5)能拼接;图像数据P(138)、P(140)、P(142)拼接之后形成的图像数据块与任何其他图像数据块都无法拼接。
因此,将图像数据P(138)、P(140)、P(142)拼接之后形成的图像数据块构成一个独立的新的图像数据块,记为Q′(Temp 1)。将图像数据P(139)、P(141)、P(143)、图像数据块Q′(3)、图像数据块Q′(5)进行拼接,形成新的图像数据块,记为Q′(Temp2)。
步骤C)判断内窥器是否已放入用户口腔,若是,则继续执行步骤D)。
步骤D)根据预设的图像模式识别算法,基于三维图像框架数据库中区块的图像特征信息,将前述新的图像数据块Q′(Temp 1)、Q′(Temp 2)与所述三维图像框架数据库中区块的图像进行匹配。
步骤E)匹配之后,若前述新的图像数据块Q′(Temp 1)无法获得与三维图像框架数据库中区块的映射关系,前述新的图像数据块Q′(Temp 2)获得与三维图像框架数据库中区块的映射关系,本次内窥操作开始以来,前述新的图像数据块Q′(Temp 2)是目前第2个获得与三维图像框架数据库中区块的映射关系的图像数据块(也即T=2),则系统对图像数据块的处理方法如下:
根据映射关系,确定前述新的图像数据块Q′(Temp 2)对应的区块,并确定该区块在保存的用户的三维图像轮廓中的位置,并将前述新的图像数据块Q′(Temp 2)重构于用户的三 维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据重构后的三维图像数据,更新当前保存的用户的三维图像模型。
并且,对三维图像框架数据库中已确定获得与前述新的图像数据块Q′(Temp 2)的映射关系的各个区块的区块编号予以记录,例如,已确定获得映射关系的区块包括:右上5牙齿近中邻面区块(编号07.061.001)、右上5牙齿颊侧区块(编号07.062.001)。
例如,本次匹配处理之前,已有记录的获得映射关系的区块的编号信息为,07.045.001、07.046.001、07.047.001、07.050.001,则加上本次处理之后记录的区块的编号信息,目前记录的所有本次内窥器开机以来,已确定获得与拼接后的图像数据块的映射关系的各个区块的编号信息为,编号07.045.001、07.046.001、07.047.001、07.050.001,07.061.001、07.062.001。
对于前述新的图像数据块Q′(Temp 2),系统记为Q°(2),并记入已获得与三维图像框架数据库中区块的映射关系的图像数据块序列{Q°(1)、Q°(2)}。
对于当前保存的图像数据块,即拼接后的图像数据块中除已确定获得映射关系的其它图像数据块Q′(1)、Q′(2)、Q′(4)、Q′(6)、…….、Q′(18),以及新的图像数据块Q′(Temp 1),按包括的图像数据的个数或面积从大到小顺序进行排序,获得排序后的当前最新的图像数据块序列{Q′(1)、…、Q′(m)、…、Q′(17),M=17}并予以保存。
根据第一预设展示策略,对更新后的用户的三维图像模型以及拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体为:
若当前展示界面中展示区域的个数为3,也即d=3,从排序后的图像数据块中取前2个图像数据块Q′(1)、Q′(2),并在该3个展示区域中分别展示取出的前2个图像数据块的三维真彩图像和更新后的用户的三维图像模型,具体地:
区域1展示的内容为:更新后的用户的三维图像模型,该用户的三维图像块模块为带有用户的三维图像轮廓,以及Q°(1)和Q°(2)图像数据块的三维真彩图像。例如,Q°(1)的三维真彩曲面图像为,右上1牙齿唇侧区块右侧部分曲面、右上1牙齿远中邻面区块、右上2牙齿近中邻面区块、右上2牙齿唇侧区块及相应牙龈曲面、右上2牙齿远中邻面区块、右上3牙齿唇侧区块左侧部分曲面;Q°(2)的三维真彩曲面图像为,右上4牙齿颊侧区块右侧部分曲面、右上5牙齿近中邻面区块、右上5牙齿颊侧区块及相应牙龈。
区域2展示的内容为:图像数据块Q′(1)的三维真彩图像,例如为左下5牙齿颊侧区块的部分曲面。
区域3展示的内容为:图像数据块Q′(2)的三维真彩图像,例如为左下4牙齿咬合面区块的部分曲面。
系统返回前述步骤A)。
实施例十五:
基于上述实施例,参阅图10所示,本发明实施例十五中,图像处理装置,具体包括:
接收单元10,用于获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据,并且所述图像数据的类型为深度图像;所述采集到的用户的图像数据表示未进行拼接处理的;
处理单元11,用于针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;并分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,以及若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型;其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述区块的图像的数据包括:编号信息、图像特征信息;所述用户的三维图像模型的初始值为所述用户的三维图像轮廓;
展示单元12,用于根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示;
发送单元13,用于若确定满足预设条件,则将更新后的所述用户的三维图像模型和记录的已确定获得映射关系的区块的编号信息,以及根据预设的规则,将所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器根据更新后的所述用户的三维图像模型和已确定获得映射关系的区块的编号信息,以及所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新保存的所述三维图像框架数据库。
较佳的,所述区块的图像的位置信息包括:每个区块相互之间的空间位置关系;
所述三维图像轮廓中每一个区块的图像,为基于所述三维图像框架数据库或所述用户的三维图像模型中区块的图像的三维曲面形状,包括预设的单一颜色和单一纹理的图像。
较佳的,分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配之前,处理单元11进一步用于:
根据拼接后的图像数据块和预设的图像模式识别算法,确定所述内窥器已放入口腔。
较佳的,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,展示 单元12具体用于:
根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;
若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示更新后的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示更新后的所述用户的三维图像模型和排序后的图像数据块。
较佳的,处理单元11进一步用于:
若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,记录已确定获得映射关系的区块的编号信息。
较佳的,处理单元11进一步用于若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中有至少一个已获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则展示单元12进一步用于:
根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
较佳的,根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,展示单元12具体用于:
根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;
若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示当前保存的所述用户的三维图像模型和所述取出的前d-1个图像数据块;
若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示当前保存的所述用户的三维图像模型和排序后的图像数据块。
较佳的,处理单元11进一步用于若确定未获得包含所述图像数据的图像数据块与所述 三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则展示单元12进一步用于:
根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
较佳的,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,展示单元12具体用于:
若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块。
较佳的,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,处理单元11进一步用于:
确定所述图像数据的个数大于预设阈值。
较佳的,所述接收单元10进一步用于在所述展示单元12根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示之后,返回执行获取内窥器采集到的用户的图像数据。
实施例十六:
基于上述实施例,参阅图11所示,本发明实施例十六中,图像处理装置,具体包括:
接收单元20,用于获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据,并且所述图像数据的类型为深度图像;所述采集到的用户的图像数据表示未进行拼接处理的;
处理单元21,用于针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;并分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配;其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述区块的图像的数据包括:编号信息、图像特征信息;
展示单元22,用于若处理单元确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后 的图像数据块的全部或部分进行展示;
发送单元23,用于若确定满足预设条件,则根据预设的规则,将所述拼接后的图像数据块的部分或全部,上传到云端服务器,以使云端服务器根据所述拼接后的图像数据块的部分或全部,更新保存的所述三维图像框架数据库。
较佳的,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,展示单元22具体用于:
若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;
确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块;
所述接收单元20进一步用于在所述展示单元22根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示之后,返回执行获取内窥器采集到的用户的图像数据。
较佳的,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,处理单元21进一步用于:
确定所述图像数据的个数大于预设阈值。
值得说明的是,本发明实施例中,上述对于实施例十五和实施例十六中的图像处理装置,其中接收单元、处理单元、展示单元和发送单元,都可以集成在一个用户终端中,例如都集成在手机中,当然,也可以分开,例如,对于带有手柄的内窥器,可以将接收单元、处理单元和发送单元集成在内窥器的手柄中,展示单元集成在手机中;又或者,将接收单元,以及处理单元的部分功能集成在内窥器的手柄中,处理单元的其它功能、展示单元和发送单元集成在手机中,在实际实现时,都是可以的,本发明实施例中并不进行限制。并且,内窥器将图像数据发送给手机,在手机上执行三维曲面图像拼接、处理,然后把处理后的三维真彩图像在手机上进行展示,这个都在本地进行,无需通过互联网连接云端,具体在实际实现时也可以在手机芯片内置引擎来实现,当然,也可以把内置ANN引擎的芯片设置在内窥器的手柄内来实现。
实施例十七:
基于上述实施例,本发明实施例中的图像处理方法,在实际工程实现时,可以包括以下执行操作:
步骤1):获取内窥器采集到的图像数据,以及获得拼接后的图像数据块。
步骤2):基于现有口腔图像数据库以及步骤1)中的图像数据块,构建人类口腔内窥 全景标签体系,即三维图像框架数据库,其中三维图像框架数据库的标签体系至少包括区块的编号信息、相对位置信息、图像信息等。
步骤3):基于步骤2)的标签体系,通过深度学习和/或强化学习等训练方法对ANN进行训练,得到最新版本的执行ANN,其中,ANN表示具备特定架构及权值的多层神经网络。
步骤4):把最新版本的执行ANN置入芯片中以承担图像匹配处理功能,例如,寒武纪或麒麟970NPU等AI芯片。
步骤5):对内窥器放入口腔后采集到的图像数据进行拼接、匹配、重构等处理,由此实现用户自助式的多初始点内窥图像三维真彩展示,即可以获得用户口腔的三维图像,进而可以基于用户口腔的三维图像,对用户当前的口腔健康状况进行人工智能检查。
步骤6):对于超出现有标签体系的图像数据,一是对图像数据进行拼接操作,获得拼接后的图像数据块,以满足用户内窥需求,二是上传到云端服务器并可以分发给牙科专家以更新三维图像框架数据库的标签体系。随后,全系统得到最新版本的口腔内窥全景标签体系,即更新后的三维图像框架数据库。之后,重新执行步骤2)。
本领域内的技术人员应明白,本发明的实施例可提供为方法、系统、或计算机程序产品。因此,本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROA、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本发明的优选实施例,但本领域内的技术人员一旦得知了基本创造性概念,则可对这些实施例作出另外的变更和修改。所以,所附权利要求意欲解释为包括优选实施例以及落入本发明范围的所有变更和修改。
显然,本领域的技术人员可以对本发明实施例进行各种改动和变型而不脱离本发明实施例的精神和范围。这样,倘若本发明实施例的这些修改和变型属于本发明权利要求及其等同技术的范围之内,则本发明也意图包含这些改动和变型在内。
Claims (30)
- 一种图像处理方法,其特征在于,包括:步骤A:获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;步骤B:针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;步骤C:分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;步骤D:若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型,以及,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示;其中,所述用户的三维图像模型的初始值为所述用户的三维图像轮廓。
- 如权利要求1所述的方法,其特征在于,所述区块的图像的位置信息包括:每个区块相互之间的空间位置关系;所述三维图像轮廓中每一个区块的图像,为基于所述三维图像框架数据库或所述用户的三维图像模型中区块的图像的三维曲面形状,包括预设的单一颜色和单一纹理的图像。
- 如权利要求1所述的方法,其特征在于,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体包括:根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示更新后的所述用户的三维图像模型和所述取出的前d-1个图像数据块;若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1 相同,并分别在调整后的当前展示界面中的展示区域中,依次展示更新后的所述用户的三维图像模型和排序后的图像数据块。
- 如权利要求1所述的方法,其特征在于,进一步包括:若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,记录已确定获得映射关系的区块的编号信息。
- 如权利要求1所述的方法,其特征在于,若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中有至少一个已获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则所述步骤D进一步包括:根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
- 如权利要求5所述的方法,其特征在于,根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,具体包括:根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示当前保存的所述用户的三维图像模型和所述取出的前d-1个图像数据块;若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示当前保存的所述用户的三维图像模型和排序后的图像数据块。
- 如权利要求1所述的方法,其特征在于,若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则所述步骤D进一步包括:根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
- 如权利要求7所述的方法,其特征在于,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,具体包括:若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块。
- 如权利要求1所述的方法,其特征在于,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,进一步包括:确定所述图像数据的个数大于预设阈值。
- 如权利要求1-9任一项所述的方法,其特征在于,进一步包括:若确定满足预设条件,则将更新后的所述用户的三维图像模型和记录的已确定获得映射关系的区块的编号信息,以及根据预设的规则,将所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器根据更新后的所述用户的三维图像模型和已确定获得映射关系的区块的编号信息,以及所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新保存的所述三维图像框架数据库。
- 如权利要求1所述的方法,其特征在于,进一步包括:执行步骤D之后,返回执行所述步骤A。
- 一种图像处理方法,其特征在于,包括:步骤A:获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;步骤B:针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;步骤C:分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;步骤D:若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
- 如权利要求12所述的方法,其特征在于,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,具体包括:若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图 像数据的个数,将所述拼接后的图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块;所述方法进一步包括:执行步骤D之后,返回执行所述步骤A。
- 如权利要求12所述的方法,其特征在于,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,进一步包括:确定所述图像数据的个数大于预设阈值。
- 如权利要求12-14任一项所述的方法,其特征在于,进一步包括:若确定满足预设条件,则根据预设的规则,将所述拼接后的图像数据块的部分或全部,上传到云端服务器,以使云端服务器根据所述拼接后的图像数据块的部分或全部,更新保存的所述三维图像框架数据库。
- 一种图像处理装置,其特征在于,包括:接收单元,用于获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;处理单元,用于针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;并分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配,以及若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,确定所述至少一个包含所述图像数据的图像数据块对应的区块在保存的用户的三维图像轮廓中的位置,并将所述至少一个包含所述图像数据的图像数据块重构于所述用户的三维图像轮廓中对应的确定的位置上,获得重构后的三维图像数据,根据所述重构后的三维图像数据,更新当前保存的所述用户的三维图像模型;其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;所述用户的三维图像模型的初始值为所述用户的三维图像轮廓;展示单元,用于根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
- 如权利要求16所述的装置,其特征在于,所述区块的图像的位置信息包括:每个区块相互之间的空间位置关系;所述三维图像轮廓中每一个区块的图像,为基于所述三维图像框架数据库或所述用户的三维图像模型中区块的图像的三维曲面形状,包括预设的单一颜色和单一纹理的图像。
- 如权利要求16所述的装置,其特征在于,根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,展示单元具体用于:根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示更新后的所述用户的三维图像模型和所述取出的前d-1个图像数据块;若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示更新后的所述用户的三维图像模型和排序后的图像数据块。
- 如权利要求16所述的装置,其特征在于,处理单元进一步用于:若确定获得至少一个包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,则根据所述映射关系,记录已确定获得映射关系的区块的编号信息。
- 如权利要求16所述的装置,其特征在于,处理单元进一步用于若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中有至少一个已获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则展示单元进一步用于:根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示。
- 如权利要求20所述的装置,其特征在于,根据第一预设展示策略,对当前保存的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示,展示单元具体用于:根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d和所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的个数n,并判断n与d-1取值的大小;若n不小于d-1,则从排序后的图像数据块中取出前d-1个图像数据块,并分别在所述d个展示区域中,依次展示当前保存的所述用户的三维图像模型和所述取出的前d-1个图 像数据块;若n小于d-1,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与n+1相同,并分别在调整后的当前展示界面中的展示区域中,依次展示当前保存的所述用户的三维图像模型和排序后的图像数据块。
- 如权利要求16所述的装置,其特征在于,处理单元进一步用于若确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则展示单元进一步用于:根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
- 如权利要求22所述的装置,其特征在于,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,展示单元具体用于:若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块。
- 如权利要求16所述的装置,其特征在于,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,处理单元进一步用于:确定所述图像数据的个数大于预设阈值。
- 如权利要求16-24任一项所述的装置,其特征在于,进一步包括:发送单元,用于若确定满足预设条件,则将更新后的所述用户的三维图像模型和记录的已确定获得映射关系的区块的编号信息,以及根据预设的规则,将所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,上传到云端服务器,以使云端服务器根据更新后的所述用户的三维图像模型和已确定获得映射关系的区块的编号信息,以及所述拼接后的图像数据块中除已确定获得与所述三维图像框架数据库中区块的映射关系的图像数据块之外的部分或全部图像数据块,更新保存的所述三维图像框架数据库。
- 如权利要求16所述的装置,其特征在于,所述接收单元进一步用于在所述展示单元根据第一预设展示策略,对更新后的所述用户的三维图像模型以及所述拼接后的图像数据块中除已确定获得映射关系的其它图像数据块的全部或部分进行展示之后,返回执行获取内窥器采集到的用户的图像数据。
- 一种图像处理装置,其特征在于,包括:接收单元,用于获取内窥器采集到的用户的图像数据;其中,所述图像数据至少包括所述内窥器中摄像单元拍摄到的图像数据;处理单元,用于针对所述图像数据,和/或所述图像数据与保存的图像数据块进行拼接,获得拼接后的图像数据块,其中,所述图像数据块包括至少一个图像数据;并分别将所述拼接后的图像数据块中包含所述图像数据的图像数据块与保存的三维图像框架数据库中区块的图像进行匹配;其中,所述三维图像框架数据库存储有将三维图像框架图像划分出的区块的图像的数据以及每个区块的图像的位置信息;展示单元,用于若处理单元确定未获得包含所述图像数据的图像数据块与所述三维图像框架数据库中区块的映射关系,并所述拼接后的图像数据块中没有获得与所述三维图像框架数据库中区块的映射关系的图像数据块,则根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示。
- 如权利要求27所述的装置,其特征在于,根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示,展示单元具体用于:若所述拼接后的图像数据块的个数为a,根据图像数据块的面积或图像数据块包括图像数据的个数,将所述拼接后的图像数据块从大到小进行排序;确定当前展示界面中的展示区域的个数d,判断a与d取值的大小,若a不小于d,则从排序后的a个图像数据块中取出前d个图像数据块,并分别在所述d个展示区域中,依次展示取出的所述前d个图像数据块,若a小于d,则调整当前展示界面中的展示区域的个数,使得展示区域的个数与a相同,并分别在调整后的当前展示界面中的展示区域中,依次展示所述a个图像数据块;所述接收单元进一步用于在所述展示单元根据第二预设展示策略,对所述拼接后的图像数据块的全部或部分进行展示之后,返回执行获取内窥器采集到的用户的图像数据。
- 如权利要求27所述的装置,其特征在于,针对所述图像数据以及所述图像数据与保存的图像数据块进行拼接之前,处理单元进一步用于:确定所述图像数据的个数大于预设阈值。
- 如权利要求27-29任一项所述的装置,其特征在于,进一步包括:发送单元,用于若确定满足预设条件,则根据预设的规则,将所述拼接后的图像数据块的部分或全部,上传到云端服务器,以使云端服务器根据所述拼接后的图像数据块的部分或全部,更新保存的所述三维图像框架数据库。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US16/759,377 US11107188B2 (en) | 2017-11-01 | 2018-10-17 | Image processing method and device |
| EP18872141.9A EP3706074B1 (en) | 2017-11-01 | 2018-10-17 | Image processing method and apparatus |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711058265.7 | 2017-11-01 | ||
| CN201711058265.7A CN107909609B (zh) | 2017-11-01 | 2017-11-01 | 一种图像处理方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019085760A1 true WO2019085760A1 (zh) | 2019-05-09 |
Family
ID=61843270
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/110709 Ceased WO2019085760A1 (zh) | 2017-11-01 | 2018-10-17 | 一种图像处理方法及装置 |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US11107188B2 (zh) |
| EP (1) | EP3706074B1 (zh) |
| CN (1) | CN107909609B (zh) |
| TW (1) | TWI712992B (zh) |
| WO (1) | WO2019085760A1 (zh) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107909609B (zh) * | 2017-11-01 | 2019-09-20 | 欧阳聪星 | 一种图像处理方法及装置 |
| CN112584738B (zh) * | 2018-08-30 | 2024-04-23 | 奥林巴斯株式会社 | 记录装置、图像观察装置、观察系统、观察系统的控制方法及存储介质 |
| CN111856445B (zh) * | 2019-04-11 | 2023-07-04 | 杭州海康威视数字技术股份有限公司 | 一种目标检测方法、装置、设备及系统 |
| CN110811550A (zh) * | 2019-10-16 | 2020-02-21 | 杨扬 | 基于深度图像的牙齿成像系统及成像方法 |
| US12033740B2 (en) * | 2021-12-04 | 2024-07-09 | Vasileios K. Papaioannou | Systems and methods for robust and automatic face de-identification for CT and MRI |
| CN115205453B (zh) * | 2022-06-23 | 2026-01-30 | 京东方科技集团股份有限公司 | 三维模型获取方法、装置、电子设备、芯片和存储介质 |
| CN119770210B (zh) * | 2025-03-07 | 2025-12-12 | 四川大学 | 一种无牙颌种植印模的辅助扫描和数据校正装置和方法 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110100340A (ko) * | 2010-03-04 | 2011-09-14 | 주식회사 사이버메드 | 가상 엔도스코프와 엔도스코프를 정합하는 방법 |
| CN103356155A (zh) * | 2013-06-24 | 2013-10-23 | 清华大学深圳研究生院 | 虚拟内窥镜辅助的腔体病灶检查系统 |
| CN104885120A (zh) * | 2012-12-31 | 2015-09-02 | 基文影像公司 | 用于显示图像流的系统和方法 |
| CN105303615A (zh) * | 2015-11-06 | 2016-02-03 | 中国民航大学 | 一种图像二维拼接与三维表面重建的组合方法 |
| CN107295227A (zh) * | 2017-05-27 | 2017-10-24 | 欧阳聪星 | 一种口腔内窥器 |
| CN107909609A (zh) * | 2017-11-01 | 2018-04-13 | 欧阳聪星 | 一种图像处理方法及装置 |
Family Cites Families (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030158786A1 (en) * | 1999-02-26 | 2003-08-21 | Skyline Software Systems, Inc. | Sending three-dimensional images over a network |
| US7330577B2 (en) * | 2004-01-27 | 2008-02-12 | Densys Ltd. | Three-dimensional modeling of the oral cavity by projecting a two-dimensional array of random patterns |
| CA2824665C (en) * | 2011-01-11 | 2016-11-01 | Nobuchika Urakabe | Intraoral video camera and display system |
| CN103392191B (zh) * | 2011-02-22 | 2016-09-28 | 3M创新有限公司 | 混合拼接 |
| WO2014100950A1 (en) * | 2012-12-24 | 2014-07-03 | Carestream Health, Inc. | Three-dimensional imaging system and handheld scanning device for three-dimensional imaging |
| EP2964139B2 (en) * | 2013-03-08 | 2025-02-12 | 3Shape A/S | Visualising a 3d dental restoration on a 2d image |
| EP3005232A4 (en) * | 2013-05-29 | 2017-03-15 | Kang-Huai Wang | Reconstruction of images from an in vivo multi-camera capsule |
| US9510757B2 (en) * | 2014-05-07 | 2016-12-06 | Align Technology, Inc. | Identification of areas of interest during intraoral scans |
| CN106663327B (zh) * | 2014-08-27 | 2020-01-24 | 锐珂牙科技术顶阔有限公司 | 3-d表面的自动重新拼接 |
| US10453269B2 (en) * | 2014-12-08 | 2019-10-22 | Align Technology, Inc. | Intraoral scanning using ultrasound and optical scan data |
| DE102015212806A1 (de) * | 2015-07-08 | 2017-01-12 | Sirona Dental Systems Gmbh | System und Verfahren zum Scannen von anatomischen Strukturen und zum Darstellen eines Scanergebnisses |
| CN105654548B (zh) * | 2015-12-24 | 2018-10-16 | 华中科技大学 | 一种基于大规模无序图像的多起点增量式三维重建方法 |
| CN105912874B (zh) * | 2016-04-29 | 2024-04-19 | 青岛大学附属医院 | 基于dicom医学图像构建的肝脏三维数据库系统 |
| US10410365B2 (en) * | 2016-06-02 | 2019-09-10 | Verily Life Sciences Llc | System and method for 3D scene reconstruction with dual complementary pattern illumination |
| WO2018101936A1 (en) * | 2016-11-30 | 2018-06-07 | CapsoVision, Inc. | Method and apparatus for image stitching of images captured using a capsule camera |
| US10695150B2 (en) * | 2016-12-16 | 2020-06-30 | Align Technology, Inc. | Augmented reality enhancements for intraoral scanning |
| GB201708520D0 (en) * | 2017-05-27 | 2017-07-12 | Dawood Andrew | A method for reducing artefact in intra oral scans |
| CN107644454B (zh) * | 2017-08-25 | 2020-02-18 | 北京奇禹科技有限公司 | 一种图像处理方法及装置 |
-
2017
- 2017-11-01 CN CN201711058265.7A patent/CN107909609B/zh active Active
-
2018
- 2018-10-17 WO PCT/CN2018/110709 patent/WO2019085760A1/zh not_active Ceased
- 2018-10-17 US US16/759,377 patent/US11107188B2/en active Active
- 2018-10-17 EP EP18872141.9A patent/EP3706074B1/en active Active
- 2018-10-25 TW TW107137689A patent/TWI712992B/zh active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110100340A (ko) * | 2010-03-04 | 2011-09-14 | 주식회사 사이버메드 | 가상 엔도스코프와 엔도스코프를 정합하는 방법 |
| CN104885120A (zh) * | 2012-12-31 | 2015-09-02 | 基文影像公司 | 用于显示图像流的系统和方法 |
| CN103356155A (zh) * | 2013-06-24 | 2013-10-23 | 清华大学深圳研究生院 | 虚拟内窥镜辅助的腔体病灶检查系统 |
| CN105303615A (zh) * | 2015-11-06 | 2016-02-03 | 中国民航大学 | 一种图像二维拼接与三维表面重建的组合方法 |
| CN107295227A (zh) * | 2017-05-27 | 2017-10-24 | 欧阳聪星 | 一种口腔内窥器 |
| CN107909609A (zh) * | 2017-11-01 | 2018-04-13 | 欧阳聪星 | 一种图像处理方法及装置 |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP3706074A4 |
Also Published As
| Publication number | Publication date |
|---|---|
| US11107188B2 (en) | 2021-08-31 |
| TWI712992B (zh) | 2020-12-11 |
| EP3706074B1 (en) | 2025-02-19 |
| TW201918998A (zh) | 2019-05-16 |
| CN107909609B (zh) | 2019-09-20 |
| CN107909609A (zh) | 2018-04-13 |
| EP3706074A4 (en) | 2021-08-18 |
| US20200273148A1 (en) | 2020-08-27 |
| EP3706074A1 (en) | 2020-09-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019085760A1 (zh) | 一种图像处理方法及装置 | |
| CN107644454B (zh) | 一种图像处理方法及装置 | |
| US12070375B2 (en) | Area of interest overlay on dental site using augmented reality | |
| CN111784754B (zh) | 基于计算机视觉的牙齿正畸方法、装置、设备及存储介质 | |
| US20220409036A1 (en) | Using an intraoral mirror with an integrated camera to record dental status, and applications thereof | |
| AU2019284043B2 (en) | Identification of areas of interest during intraoral scans | |
| ES2724115T3 (es) | Interfaz gráfica de usuario para el marcado de márgenes asistido por ordenador sobre dentaduras | |
| WO2019141106A1 (zh) | 基于c/s架构的牙齿美化ar智能辅助方法及装置 | |
| US12569319B2 (en) | Combined face scanning and intraoral scanning | |
| US12551313B2 (en) | Dental scanning | |
| US20250339244A1 (en) | Method for acquiring a model of a dental arch | |
| US20250127597A1 (en) | Method for acquiring a model of a dental arch | |
| HK1253088B (zh) | 一种图像处理方法及装置 | |
| HK1253088A1 (zh) | 一种图像处理方法及装置 | |
| JP7695221B2 (ja) | データ生成装置、データ生成方法、およびデータ生成プログラム | |
| JP7600187B2 (ja) | データ生成装置、データ生成方法、およびデータ生成プログラム | |
| CN115471506B (zh) | 一种口扫模型的位置调整方法、存储介质和电子设备 | |
| HK1250412B (zh) | 一种图像处理方法及装置 | |
| EP4227955A1 (en) | Enhanced scanning | |
| SENKUTVAN et al. | INTRAORAL SCANNERS IN ORTHODONTICS |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18872141 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| ENP | Entry into the national phase |
Ref document number: 2018872141 Country of ref document: EP Effective date: 20200602 |