WO2022166506A1 - 图像处理方法和装置 - Google Patents

图像处理方法和装置 Download PDF

Info

Publication number
WO2022166506A1
WO2022166506A1 PCT/CN2022/070055 CN2022070055W WO2022166506A1 WO 2022166506 A1 WO2022166506 A1 WO 2022166506A1 CN 2022070055 W CN2022070055 W CN 2022070055W WO 2022166506 A1 WO2022166506 A1 WO 2022166506A1
Authority
WO
WIPO (PCT)
Prior art keywords
slope
line
edge
image
straight line
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
Application number
PCT/CN2022/070055
Other languages
English (en)
French (fr)
Inventor
杨玫
周芳汝
安山
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jingdong Century Trading Co Ltd
Beijing Wodong Tianjun Information Technology Co Ltd
Original Assignee
Beijing Jingdong Century Trading Co Ltd
Beijing Wodong Tianjun Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Jingdong Century Trading Co Ltd, Beijing Wodong Tianjun Information Technology Co Ltd filed Critical Beijing Jingdong Century Trading Co Ltd
Priority to US18/263,005 priority Critical patent/US12586147B2/en
Priority to EP22748785.7A priority patent/EP4270316A4/en
Publication of WO2022166506A1 publication Critical patent/WO2022166506A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06T3/02—Affine transformations
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • 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
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—Two-dimensional [2D] image generation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20224—Image subtraction

Definitions

  • the present disclosure relates to the field of information processing, and in particular, to an image processing method and device.
  • image display is widely used as a means of publicity. For example, by displaying items such as floors and carpets in different decoration scenarios, the user experience can be effectively improved.
  • an image processing method comprising: performing line detection on a target boundary in a scene image to obtain multiple candidate straight lines;
  • the candidate line with the largest number of overlapping pixels of the boundary is used as the first reference line;
  • the minimum circumscribed rectangle of the target is generated in the scene image;
  • the transformation area is generated, wherein the four vertices of the minimum circumscribed rectangle are located in the transformation area.
  • the slopes of the first edge and the opposite second edge of the transformation area are related to the slope of the first reference line, and the slopes of the third edge and the opposite fourth edge of the transformation area are related to the slope of the first reference line. correlating the slopes of the outer angle bisectors of the first edge and the second edge; transforming a preset image into the transformation area to generate a transformed image; overlaying the target with the transformed image to update the scene image.
  • generating the transformed region includes: generating a first edge line passing through a first vertex of the minimum circumscribed rectangle and a second edge line passing through a second vertex of the minimum circumscribed rectangle, where the first vertex is at an angle
  • the angle at which the second vertex is located is a diagonal angle, and the difference between the slopes of the first edge and the second edge and the slope of the first reference line is within a first preset range; a second reference line that circumscribes the center point of the rectangle, wherein the slope of the second reference line is the slope of the bisector of the outer corners of the first edge and the second edge; generating a line passing through the third vertex of the minimum circumscribed rectangle
  • the third sideline and the fourth sideline passing through the fourth vertex of the minimum circumscribed rectangle, the angle where the third vertex is located and the angle where the fourth vertex is located are diagonal, and the slopes of the third sideline and the fourth sideline are the same as The difference between the slopes of the second reference straight line is within a second preset range;
  • the slope of the first edge when the slope of the first reference line is less than 0, the slope of the first edge is greater than the slope of the first reference line, and the slope of the second edge is smaller than the slope of the first edge - The slope of the reference line.
  • the slope of the first edge when the slope of the first reference line is greater than 0, the slope of the first edge is smaller than the slope of the first reference line, and the slope of the second edge is greater than the slope of the first edge - The slope of the reference line.
  • the difference between the slope of the first edge line and the slope of the first reference straight line is a ratio of the distance from the first vertex to the first reference straight line to a first preset parameter;
  • the difference between the slope of the second edge line and the slope of the first reference straight line is a ratio of the distance from the second vertex to the first reference straight line to a second preset parameter.
  • the slope of the third edge is smaller than the slope of the second reference line
  • the slope of the fourth edge is greater than that of the second edge Refer to the slope of the line.
  • the slope of the third edge is greater than the slope of the second reference line
  • the slope of the fourth edge is smaller than that of the second edge Refer to the slope of the line.
  • mapping a preset image into the transformation area to generate a transformed image includes: generating a minimum circumscribed rectangle of the transformed area; placing the preset image in a minimum circumscribed rectangle of the transformed area Repeated stitching in the region to generate a sample image; transform the sample image into the transformed region by affine transformation to generate a transformed image.
  • the portion of the transition image beyond the scene image is deleted.
  • an image processing apparatus including: a first processing module configured to perform line detection on a target boundary in a scene image to obtain a plurality of candidate straight lines, among the plurality of straight lines Among the candidate lines, the candidate line with the largest number of overlapping pixels with the boundary is used as the first reference line; the second processing module is configured to generate the minimum circumscribed rectangle of the target in the scene image; the transformation area generation module , is configured to generate a transformation region, wherein the four vertices of the minimum circumscribed rectangle are located on four sides of the transformation region, and the slopes of the first edge and the opposite second edge of the transformation region are the same as the first edge of the transformation region.
  • the image processing module is configured to Transforming a preset image into the transform region to generate a transform image with which the object is overlaid to update the scene image.
  • an image processing apparatus comprising: a memory configured to store instructions; a processor coupled to the memory, the processor configured to execute any of the above-mentioned instructions based on the instructions stored in the memory methods described in the examples.
  • a non-transitory computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, any of the foregoing embodiments is implemented methods involved.
  • FIG. 1 is a schematic flowchart of an image processing method according to an embodiment of the disclosure
  • FIGS. 2A-2C are schematic diagrams of object boundary extraction according to some embodiments of the present disclosure.
  • FIG. 3 is a schematic diagram of a minimum circumscribed rectangle of a target of an embodiment of the disclosure.
  • FIG. 4 is a schematic flowchart of an image processing method according to another embodiment of the disclosure.
  • FIG. 5-8 are schematic diagrams of edge generation of a transformed region according to some embodiments of the present disclosure.
  • FIG. 9 is a schematic diagram of a transformation area according to an embodiment of the present disclosure.
  • FIGS. 10A-10C are schematic diagrams of image transformation according to some embodiments of the present disclosure.
  • FIG. 12 is a schematic structural diagram of an image processing apparatus according to an embodiment of the disclosure.
  • FIG. 13 is a schematic structural diagram of an image processing apparatus according to another embodiment of the disclosure.
  • the manual labeling method has huge workload, time-consuming and high labor cost, and cannot meet the individual needs of users.
  • the present disclosure provides an image processing solution that can directly replace the target of a scene image with a sample image, so as to effectively meet the personalized needs of users.
  • FIG. 1 is a schematic flowchart of an image processing method according to an embodiment of the present disclosure. In some embodiments, the following image processing method steps are performed by an image processing apparatus.
  • step 101 line detection is performed on the target boundary in the scene image to obtain a plurality of candidate lines.
  • Hough line detection is performed on the extracted boundary to obtain a plurality of candidate lines.
  • step 102 among the plurality of candidate straight lines, the candidate straight line with the largest number of overlapping pixels with the boundary is used as the first reference straight line.
  • step 103 a minimum circumscribed rectangle of the object is generated in the scene image.
  • the minimum circumscribed rectangle of the floor is shown as the black box in Figure 3.
  • transform regions are generated.
  • the four vertices of the minimum circumscribed rectangle are located on the four edges of the transformation area, the slopes of the first edge and the opposite second edge of the transformation area are related to the slope of the first reference line, and the first edge of the transformation area is related to the slope of the first reference line.
  • the slopes of the third sideline and the opposite fourth sideline are associated with the slopes of the bisectors of the exterior angles of the first sideline and the second sideline.
  • FIG. 4 is a schematic flowchart of an image processing method according to another embodiment of the disclosure. This embodiment describes the steps of generating the transformed region. In some embodiments, the following image processing method steps are performed by an image processing apparatus.
  • step 401 a first edge line passing through the first vertex of the minimum circumscribed rectangle and a second edge line passing through the second vertex of the minimum circumscribed rectangle are generated.
  • the difference between the slopes of the line and the second edge line and the slope of the first reference straight line is within a first preset range.
  • the slope of the first reference line when the slope of the first reference line is less than 0, the slope of the first edge line is greater than the slope of the first reference line, and the slope of the second edge line is smaller than the slope of the first reference line.
  • the rectangle ABCD is the generated minimum circumscribed rectangle
  • the straight line L00 is the first reference straight line
  • the slope of the straight line L00 is k* ⁇ 0.
  • the slopes of the first edge L1 passing through the vertex A and the second edge L2 passing through the vertex D will not be completely consistent with the slope of the straight line L00, but will have a certain deviation.
  • the slope of the first edge L1 is greater than the slope of the straight line L00
  • the slope of the second edge L2 is smaller than the slope of the straight line L00 .
  • the distance from vertex A to straight line L00 is d1
  • the distance from vertex D to straight line L00 is d2
  • the slope k1 of the first edge L1 is shown in formula (1)
  • the second edge L2 The slope k2 is shown in formula (2), wherein the parameters N1 and N2 are preset parameters.
  • the slope k*>0 of the first reference straight line L00 the slopes of the first edge L1 passing through the vertex B and the second edge L2 passing through the vertex C will not be completely consistent with the slope of the straight line L00, but will have a certain deviation.
  • the slope of the first edge L1 is smaller than the slope of the straight line L00
  • the slope of the second edge L2 is greater than the slope of the straight line L00 .
  • the distance from vertex B to straight line L00 is d3
  • the distance from vertex C to straight line L00 is d4
  • the slope k1 of the first edge L1 is shown in formula (3)
  • the second edge L2 The slope k2 is shown in formula (4), wherein the parameters N1 and N2 are preset parameters.
  • step 402 a second reference straight line passing through the center point of the minimum circumscribed rectangle is generated, wherein the slope of the second reference straight line is the slope of the bisector of the outer angle of the first side line and the second side line.
  • step 403 a third edge line passing through the third vertex of the minimum circumscribed rectangle and a fourth edge line passing through the fourth vertex of the minimum circumscribed rectangle are generated.
  • the difference between the slope of the fourth edge line and the slope of the second reference straight line is within a second preset range.
  • the slope of the third edge is smaller than the slope of the second reference line
  • the slope of the fourth edge is greater than the slope of the second reference line
  • the second reference line L01 passes through the center point O of the rectangle ABCD, and the slope k**>0 of the second reference line L01.
  • the slopes of the third edge L3 passing through the vertex B and the fourth edge L4 passing through the vertex C will not be exactly the same as the slope of the straight line L01, but will have a certain deviation.
  • the slope of the third edge L3 is smaller than the slope of the straight line L01
  • the slope of the fourth edge L4 is larger than the slope of the straight line L01 .
  • the distance from vertex B to straight line L01 and the distance from vertex C to straight line L01 are both d5
  • the slope k3 of the third edge L3 is shown in formula (5)
  • the slope k4 of the fourth edge L4 is preset parameters.
  • the slope of the second reference line when the slope of the second reference line is less than 0, the slope of the third edge is greater than the slope of the second reference line, and the slope of the fourth edge is smaller than the slope of the second reference line.
  • the second reference line L01 passes through the center point O of the rectangle ABCD, and the slope k** ⁇ 0 of the second reference line L01.
  • the slopes of the third edge L3 passing through the vertex A and the fourth edge L4 passing through the vertex D will not be exactly the same as the slope of the straight line L01, but have a certain deviation.
  • the slope of the third edge L3 is greater than the slope of the straight line L01
  • the slope of the fourth edge L4 is smaller than the slope of the straight line L01 .
  • the distance from vertex A to straight line L01 and the distance from vertex D to straight line L01 are both d6, then the slope k3 of the third edge L3 is shown in formula (7), and the slope k4 of the fourth edge L4 As shown in formula (8), the parameters N3 and N4 are preset parameters.
  • a transformation region is generated using the first edge, the second edge, the third edge and the fourth edge.
  • the resulting transformed region is shown as the white line frame in FIG. 9 .
  • step 105 the preset image is transformed into the transformed region to generate a transformed image.
  • a sample image is generated by repeatedly splicing a preset image in the smallest circumscribed rectangle of the transformation region by generating a minimum circumscribed rectangle of the transformation region, and the sample image is transformed into the transformation region through affine transformation to Generate rollover images.
  • FIG. 9A For example, a preset floor pattern is shown in FIG. 9A .
  • a sample image is generated by repeatedly splicing the preset floor sample image in the smallest circumscribed rectangle of the transformed region, as shown in Figure 9B.
  • the sample image is transformed into the transformed region by affine transformation to generate a transformed image, as shown in Figure 9C.
  • the object is overlaid with the transform image in order to update the scene image.
  • the portion of the rollover image beyond the scene image is deleted.
  • FIG. 11A is the original scene image, and the transformation image is overlaid on the original scene image through the solution described in the above embodiments, so that the floor style in the original scene image will be changed, as shown in FIG. 11B .
  • the target of the scene image can be directly replaced with the sample image, thereby effectively meeting the personalized needs of the user.
  • FIG. 12 is a schematic structural diagram of an image processing apparatus according to an embodiment of the disclosure. As shown in FIG. 12 , the image processing apparatus includes a first processing module 1201 , a second processing module 1202 , a transformed region generation module 1203 and an image processing module 1204 .
  • the first processing module 1201 is configured to perform straight line detection on the target boundary in the scene image to obtain multiple candidate straight lines.
  • the candidate straight line with the largest number of overlapping pixels with the boundary is used as the first reference straight line.
  • the second processing module 1202 is configured to generate a minimum enclosing rectangle of the object in the scene image.
  • the minimum circumscribed rectangle of the floor is shown as the black box in Figure 3.
  • the transformation region generation module 1203 is configured to generate a transformation region, wherein the four vertices of the minimum circumscribed rectangle are located on the four sides of the transformation region, the slopes of the first edge line and the opposite second edge line of the transformation region and the slope of the first reference straight line Associated, the slopes of the third edge and the opposite fourth edge of the transformed region are associated with the slopes of the exterior angle bisectors of the first edge and the second edge.
  • the transform region generation module 1203 generates a first edge line passing through the first vertex of the minimum circumscribed rectangle and a second edge line passing through the second vertex of the minimum circumscribed rectangle, the corner where the first vertex is located and the corner where the second vertex is located To be diagonal, the difference between the slopes of the first edge line and the second edge line and the slope of the first reference straight line is within a first preset range.
  • the slope of the first edge line when the slope of the first reference line is less than 0, the slope of the first edge line is greater than the slope of the first reference line, and the slope of the second edge line is smaller than the slope of the first reference line.
  • the slope of the first edge can be calculated using the above formula (1)
  • the slope of the second edge can be calculated using the above formula (2).
  • the slope of the first edge line when the slope of the first reference line is greater than 0, the slope of the first edge line is smaller than the slope of the first reference line, and the slope of the second edge line is greater than the slope of the first reference line.
  • the slope of the first edge can be calculated by using the above formula (3), and the slope of the second edge can be calculated using the above formula (4).
  • the transformation region generating module 1203 generates a second reference straight line passing through the center point of the minimum circumscribed rectangle, wherein the slope of the second reference straight line is the slope of the bisector of the outer corners of the first side line and the second side line.
  • the transformation area generation module 1203 generates a third edge passing through the third vertex of the minimum circumscribed rectangle and a fourth edge passing through the fourth vertex of the minimum circumscribed rectangle, and the angle at which the third vertex is located and the angle at which the fourth vertex is located are opposite corners, The difference between the slopes of the third edge line and the fourth edge line and the slope of the second reference straight line is within a second preset range.
  • the slope of the third edge is smaller than the slope of the second reference line
  • the slope of the fourth edge is greater than the slope of the second reference line.
  • the slope of the third side line can be calculated by the above formula (5)
  • the slope of the fourth side line can be calculated by the above formula (6).
  • the slope of the third edge is greater than the slope of the second reference line
  • the slope of the fourth edge is smaller than the slope of the second reference line.
  • the slope of the third side line can be calculated by the above formula (7)
  • the slope of the fourth side line can be calculated by the above formula (8).
  • the transformation region generating module 1203 generates a transformation region using the first edge, the second edge, the third edge and the fourth edge. For example, on the basis of FIG. 8 , the resulting transformed region is shown as the white line frame in FIG. 9 .
  • the image processing module 1204 is configured to transform the preset image into the transformed region to generate the transformed image, overlay the target with the transformed image to update the scene image.
  • the image processing module 1204 generates a sample image by repeatedly splicing the preset image in the minimum circumscribing rectangle of the transformation region by generating the minimum enclosing rectangle of the transformation region, and transforms the sample image into the transformation region through affine transformation. area to generate a rollover image.
  • FIG. 13 is a schematic structural diagram of an image processing apparatus according to another embodiment of the disclosure. As shown in FIG. 13 , the image processing apparatus includes a memory 1301 and a processor 1302 .
  • the memory 1301 is used to store instructions, and a processor 1302 is coupled to the memory 1301, and the processor 1302 is configured to implement the method involved in any of the embodiments in FIG. 1 or FIG. 4 based on the instructions stored in the memory.
  • the image processing apparatus further includes a communication interface 1303 for exchanging information with other devices. Meanwhile, the image processing apparatus further includes a bus 1304 , the processor 1302 , the communication interface 1303 , and the memory 1301 communicate with each other through the bus 1304 .
  • the memory 1301 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory. Memory 1301 may also be a memory array.
  • the storage 1301 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules.
  • processor 1302 may be a central processing unit (CPU), or may be an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present disclosure.
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • the present disclosure also relates to a non-transitory computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the instructions are executed by a processor to implement the method involved in any of the embodiments in FIG. 1 or FIG. 4 .
  • the functional unit modules described above may be implemented as a general-purpose processor, a programmable logic controller (Programmable Logic Controller, PLC for short), a digital signal processor ( Digital Signal Processor (referred to as: DSP), Application Specific Integrated Circuit (referred to as: ASIC), Field-Programmable Gate Array (referred to as: FPGA) or other programmable logic devices, discrete gates or transistors Logic devices, discrete hardware components, or any suitable combination thereof.
  • a programmable logic controller Programmable Logic Controller, PLC for short
  • DSP Digital Signal Processor
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

一种图像处理方法和装置。图像处理方法包括：对场景图像中的目标边界进行直线检测，以得到多个候选直线（101）；在多个候选直线中，将与边界的重叠像素数量最多的候选直线作为第一参照直线（102）；在场景图像中生成目标的最小外接矩形（103）；生成变换区域（104），其中最小外接矩形的四个顶点位于变换区域的四条边线上，变换区域的第一边线和相对的第二边线的斜率与第一参照直线的斜率相关联，变换区域的第三边线和相对的第四边线的斜率与第一边线和第二边线的外角平分线的斜率相关联；将预设图像变换到变换区域中，以生成变换图像（105）；利用变换图像覆盖目标，以便更新场景图像（106）。

Description

图像处理方法和装置
相关申请的交叉引用
本公开是以CN申请号为202110175113.5，申请日为2021年2月7日的申请为基础，并主张其优先权，该CN申请的公开内容在此作为整体引入本公开中。
技术领域
本公开涉及信息处理领域，特别涉及一种图像处理方法和装置。
背景技术
目前，将图像展示作为宣传手段以得到了广泛应用。例如通过展示在不同装修场景下的地板、地毯等物品，可有效提升用户的体验。
发明内容
根据本公开实施例的第一方面，提供一种图像处理方法，包括：对场景图像中的目标边界进行直线检测，以得到多个候选直线；在所述多个候选直线中，将与所述边界的重叠像素数量最多的候选直线作为第一参照直线；在所述场景图像中生成所述目标的最小外接矩形；生成变换区域，其中所述最小外接矩形的四个顶点位于所述变换区域的四条边线上，所述变换区域的第一边线和相对的第二边线的斜率与所述第一参照直线的斜率相关联，所述变换区域的第三边线和相对的第四边线的斜率与所述第一边线和第二边线的外角平分线的斜率相关联；将预设图像变换到所述变换区域中，以生成变换图像；利用所述变换图像覆盖所述目标，以便更新所述场景图像。
在一些实施例中，生成变换区域包括：生成经过所述最小外接矩形的第一顶点的第一边线和经过所述最小外接矩形的第二顶点的第二边线，所述第一顶点所在角和所述第二顶点所在角为对角，所述第一边线和所述第二边线的斜率与所述第一参照直线的斜率之差在第一预设范围内；生成经过所述最小外接矩形的中心点的第二参照直线，其中所述第二参照直线的斜率为所述第一边线和第二边线的外角平分线的斜率；生成经过所述最小外接矩形的第三顶点的第三边线和经过所述最小外接矩形的第四顶点的第四边线，所述第三顶点所在角和第四顶点所在角为对角，所述第三边线和所述第四边线的斜率与所述第二参照直线的斜率之差在第二预设范围内；利用所述第一 边线、第二边线、第三边线和第四边线生成变换区域。
在一些实施例中，在所述第一参照直线的斜率小于0的情况下，所述第一边线的斜率大于所述第一参照直线的斜率，所述第二边线的斜率小于所述第一参照直线的斜率。
在一些实施例中，在所述第一参照直线的斜率大于0的情况下，所述第一边线的斜率小于所述第一参照直线的斜率，所述第二边线的斜率大于所述第一参照直线的斜率。
在一些实施例中，所述第一边线的斜率与所述第一参照直线的斜率之差为所述第一顶点到所述第一参照直线的距离和第一预设参数的比值；所述第二边线的斜率与所述第一参照直线的斜率之差为所述第二顶点到所述第一参照直线的距离和第二预设参数的比值。
在一些实施例中，在所述第二参照直线的斜率大于0的情况下，所述第三边线的斜率小于所述第二参照直线的斜率，所述第四边线的斜率大于所述第二参照直线的斜率。
在一些实施例中，在所述第二参照直线的斜率小于0的情况下，所述第三边线的斜率大于所述第二参照直线的斜率，所述第四边线的斜率小于所述第二参照直线的斜率。
在一些实施例中，所述第三边线的斜率与所述第二参照直线的斜率之差为所述第三顶点到所述第二参照直线的距离和第三预设参数的比值；所述第四边线的斜率与所述第二参照直线的斜率之差为所述第四顶点到所述第二参照直线的距离和第四预设参数的比值。
在一些实施例中，将预设图像映射变换到所述变换区域中，以生成变换图像包括：生成所述变换区域的最小外接矩形；将所述预设图像在所述变换区域的最小外接矩形中重复拼接，以生成样本图像；通过仿射变换将所述样本图像变换到所述变换区域中，以生成变换图像。
在一些实施例中，在利用所述变换图像覆盖所述目标后，将所述变换图像中超出所述场景图像的部分删除。
根据本公开实施例的第二方面，提供一种图像处理装置，包括：第一处理模块，被配置为对场景图像中的目标边界进行直线检测，以得到多个候选直线，在所述多个候选直线中，将与所述边界的重叠像素数量最多的候选直线作为第一参照直线；第二 处理模块，被配置为在所述场景图像中生成所述目标的最小外接矩形；变换区域生成模块，被配置为生成变换区域，其中所述最小外接矩形的四个顶点位于所述变换区域的四条边线上，所述变换区域的第一边线和相对的第二边线的斜率与所述第一参照直线的斜率相关联，所述变换区域的第三边线和相对的第四边线的斜率与所述第一边线和第二边线的外角平分线的斜率相关联；图像处理模块，被配置为将预设图像变换到所述变换区域中，以生成变换图像，利用所述变换图像覆盖所述目标，以便更新所述场景图像。
根据本公开实施例的第三方面，提供一种图像处理装置，包括：存储器，被配置为存储指令；处理器，耦合到存储器，处理器被配置为基于存储器存储的指令执行实现如上述任一实施例所述的方法。
根据本公开实施例的第四方面，提供一种非瞬态计算机可读存储介质，其中，非瞬态计算机可读存储介质存储有计算机指令，指令被处理器执行时实现如上述任一实施例涉及的方法。
通过以下参照附图对本公开的示例性实施例的详细描述，本公开的其它特征及其优点将会变得清楚。
附图说明
为了更清楚地说明本公开实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本公开的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本公开一个实施例的图像处理方法的流程示意图；
图2A-图2C为本公开一些实施例的目标边界提取示意图；
图3为本公开一个实施例的目标的最小外接矩形的示意图；
图4为本公开另一个实施例的图像处理方法的流程示意图；
图5-图8为本公开一些实施例的变换区域边线生成示意图；
图9为本公开一个实施例的变换区域示意图；
图10A-图10C为本公开一些实施例的图像变换示意图；
图11A-图11B为本公开一些实施例的目标呈现示意图；
图12为本公开一个实施例的图像处理装置的结构示意图；
图13为本公开另一个实施例的图像处理装置的结构示意图。
具体实施方式
下面将结合本公开实施例中的附图，对本公开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公开一部分实施例，而不是全部的实施例。以下对至少一个示例性实施例的描述实际上仅仅是说明性的，决不作为对本公开及其应用或使用的任何限制。基于本公开中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本公开保护的范围。
除非另外具体说明，否则在这些实施例中阐述的部件和步骤的相对布置、数字表达式和数值不限制本公开的范围。
同时，应当明白，为了便于描述，附图中所示出的各个部分的尺寸并不是按照实际的比例关系绘制的。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为授权说明书的一部分。
在这里示出和讨论的所有示例中，任何具体值应被解释为仅仅是示例性的，而不是作为限制。因此，示例性实施例的其它示例可以具有不同的值。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步讨论。
发明人注意到，在相关技术中，通过人工手动标注的方式对场景图像中诸如地板的目标进行更换。然而人工手动标注方式的工作量巨大，耗时长且人工成本高，无法满足用户的个性化需求。
据此，本公开提供一种图像处理方案，能够直接将场景图像的目标更换为样本图像，从而有效满足用户的个性化需求。
图1为本公开一个实施例的图像处理方法的流程示意图。在一些实施例中，下列的图像处理方法步骤由图像处理装置执行。
在步骤101，对场景图像中的目标边界进行直线检测，以得到多个候选直线。
例如，为了对如图2A所示的图像中的地板区域进行处理，首先利用地板区域的前景掩膜mask将场景图像中的地板区域提取出来，并将非地板区域像素值置为0，如图2B所示。接下来，利用Canny算子提取地板区域的边界，如图2C所示。
然后，对提取到的边界进行霍夫(Hough)直线检测，以得到多个候选直线。
在步骤102，在多个候选直线中，将与边界的重叠像素数量最多的候选直线作为第一参照直线。
在步骤103，在场景图像中生成目标的最小外接矩形。
例如，在图2A中所示的图像中，地板的最小外接矩形如图3中的黑色方框所示。
在步骤104，生成变换区域。
这里需要说明的是，最小外接矩形的四个顶点位于变换区域的四条边线上，变换区域的第一边线和相对的第二边线的斜率与第一参照直线的斜率相关联，变换区域的第三边线和相对的第四边线的斜率与第一边线和第二边线的外角平分线的斜率相关联。
图4为本公开另一个实施例的图像处理方法的流程示意图。该实施例描述变换区域的生成步骤。在一些实施例中，下列的图像处理方法步骤由图像处理装置执行。
在步骤401，生成经过最小外接矩形的第一顶点的第一边线和经过最小外接矩形的第二顶点的第二边线，第一顶点所在角和第二顶点所在角为对角，第一边线和第二边线的斜率与第一参照直线的斜率之差在第一预设范围内。
在一些实施例中，在第一参照直线的斜率小于0的情况下，第一边线的斜率大于第一参照直线的斜率，第二边线的斜率小于第一参照直线的斜率。
在一些实施例中，如图5所示，矩形ABCD为所生成的最小外接矩形，直线L00为第一参照直线，直线L00的斜率k*<0。考虑到视觉效果，经过顶点A的第一边线L1和经过顶点D的第二边线L2的斜率不会与直线L00的斜率完全一致，而是会具有一定的偏差。例如，如图5所示，第一边线L1的斜率大于直线L00的斜率，第二边线L2的斜率小于直线L00的斜率。
例如，如图5所示，顶点A到直线L00的距离为d1，顶点D到直线L00的距离为d2，则第一边线L1的斜率k1如公式(1)所示，第二边线L2的斜率k2如公式(2)所示，其中参数N1和N2为预设参数。
k1＝k*+d1/N1                         (1)
k2＝k*-d2/N2                         (2)
在一些实施例中，如图6所示，第一参照直线L00的斜率k*>0。考虑到视觉效果，经过顶点B的第一边线L1和经过顶点C的第二边线L2的斜率不会与直线L00的斜率完全一致，而是会具有一定的偏差。例如，如图6所示，第一边线L1的斜率小于直线L00的斜率，第二边线L2的斜率大于直线L00的斜率。
例如，如图6所示，顶点B到直线L00的距离为d3，顶点C到直线L00的距离为d4，则第一边线L1的斜率k1如公式(3)所示，第二边线L2的斜率k2如公式(4)所示，其中参数N1和N2为预设参数。
k1＝k*-d3/N1                         (3)
k2＝k*+d4/N2                         (4)
在步骤402，生成经过最小外接矩形的中心点的第二参照直线，其中第二参照直线的斜率为第一边线和第二边线的外角平分线的斜率。
在步骤403，生成经过最小外接矩形的第三顶点的第三边线和经过最小外接矩形的第四顶点的第四边线，第三顶点所在角和第四顶点所在角为对角，第三边线和第四边线的斜率与第二参照直线的斜率之差在第二预设范围内。
在一些实施例中，在第二参照直线的斜率大于0的情况下，第三边线的斜率小于第二参照直线的斜率，第四边线的斜率大于第二参照直线的斜率。
在一些实施例中，如图7所示，第二参照直线L01经过矩形ABCD的中心O点，第二参照直线L01的斜率k**>0。考虑到视觉效果，经过顶点B的第三边线L3和经过顶点C的第四边线L4的斜率不会与直线L01的斜率完全一致，而是会具有一定的偏差。例如，如图7所示，第三边线L3的斜率小于直线L01的斜率，第四边线L4的斜率大于直线L01的斜率。
例如，如图7所示，顶点B到直线L01的距离和顶点C到直线L01的距离均为d5，则第三边线L3的斜率k3如公式(5)所示，第四边线L4的斜率k4如公式(6)所示，其中参数N3和N4为预设参数。
k3＝k**-d5/N3                         (5)
k4＝k**+d5/N4                         (6)
在一些实施例中，在第二参照直线的斜率小于0的情况下，第三边线的斜率大于第二参照直线的斜率，第四边线的斜率小于第二参照直线的斜率。
在一些实施例中，如图8所示，第二参照直线L01经过矩形ABCD的中心O点，第二参照直线L01的斜率k**<0。考虑到视觉效果，经过顶点A的第三边线L3和经过顶点D的第四边线L4的斜率不会与直线L01的斜率完全一致，而是会具有一定的偏差。例如，如图8所示，第三边线L3的斜率大于直线L01的斜率，第四边线L4的斜率小于直线L01的斜率。
例如，如图8所示，顶点A到直线L01的距离和顶点D到直线L01的距离均为 d6，则第三边线L3的斜率k3如公式(7)所示，第四边线L4的斜率k4如公式(8)所示，其中参数N3和N4为预设参数。
k3＝k**+d6/N3                         (7)
k4＝k**-d6/N4                         (8)
在步骤404，利用第一边线、第二边线、第三边线和第四边线生成变换区域。
例如，在图8的基础上，所得到的变换区域如图9中的白色线框所示。
返回图1。在步骤105，将预设图像变换到变换区域中，以生成变换图像。
在一些实施例中，通过生成变换区域的最小外接矩形，将预设图像在变换区域的最小外接矩形中重复拼接，以生成样本图像，并通过仿射变换将样本图像变换到变换区域中，以生成变换图像。
例如，预设的地板样图如图9A所示。通过将预设的地板样图在变换区域的最小外接矩形中重复拼接，以生成样本图像，如图9B所示。接下来，通过仿射变换将样本图像变换到变换区域中，以生成变换图像，如图9C所示。
在步骤106，利用变换图像覆盖目标，以便更新场景图像。
在一些实施例中，在利用变换图像覆盖目标后，将变换图像中超出场景图像的部分删除。
例如，图11A为原始的场景图像，通过上述实施例描述的方案在原始的场景图像上覆盖变换图像，从而将改变原始的场景图像中的地板样式，如图11B所示。
在本公开上述实施例提供的图像处理装置中，能够直接将场景图像的目标更换为样本图像，从而有效满足用户的个性化需求。
图12为本公开一个实施例的图像处理装置的结构示意图。如图12所示，图像处理装置包括第一处理模块1201、第二处理模块1202、变换区域生成模块1203和图像处理模块1204。
第一处理模块1201被配置为对场景图像中的目标边界进行直线检测，以得到多个候选直线，在多个候选直线中，将与边界的重叠像素数量最多的候选直线作为第一参照直线。
例如，为了对如图2A所示的图像中的地板区域进行处理，首先利用地板区域的前景掩膜mask将场景图像中的地板区域提取出来，并将非地板区域像素值置为0，如图2B所示。接下来，利用Canny算子提取地板区域的边界，如图2C所示。然后，对提取到的边界进行霍夫直线检测，以得到多个候选直线。
第二处理模块1202被配置为在场景图像中生成目标的最小外接矩形。
例如，在图2A中所示的图像中，地板的最小外接矩形如图3中的黑色方框所示。
变换区域生成模块1203被配置为生成变换区域，其中最小外接矩形的四个顶点位于变换区域的四条边线上，变换区域的第一边线和相对的第二边线的斜率与第一参照直线的斜率相关联，变换区域的第三边线和相对的第四边线的斜率与第一边线和第二边线的外角平分线的斜率相关联。
在一些实施例中，变换区域生成模块1203生成经过最小外接矩形的第一顶点的第一边线和经过最小外接矩形的第二顶点的第二边线，第一顶点所在角和第二顶点所在角为对角，第一边线和第二边线的斜率与第一参照直线的斜率之差在第一预设范围内。
在一些实施例中，在第一参照直线的斜率小于0的情况下，第一边线的斜率大于第一参照直线的斜率，第二边线的斜率小于第一参照直线的斜率。例如可利用上述公式(1)计算第一边线的斜率，利用上述公式(2)计算第二边线的斜率。
在一些实施例中，在第一参照直线的斜率大于0的情况下，第一边线的斜率小于第一参照直线的斜率，第二边线的斜率大于第一参照直线的斜率。例如可利用上述公式(3)计算第一边线的斜率，利用上述公式(4)计算第二边线的斜率。
接下来，变换区域生成模块1203生成经过最小外接矩形的中心点的第二参照直线，其中第二参照直线的斜率为第一边线和第二边线的外角平分线的斜率。
接下来，变换区域生成模块1203生成经过最小外接矩形的第三顶点的第三边线和经过最小外接矩形的第四顶点的第四边线，第三顶点所在角和第四顶点所在角为对角，第三边线和第四边线的斜率与第二参照直线的斜率之差在第二预设范围内。
在一些实施例中，在第二参照直线的斜率大于0的情况下，第三边线的斜率小于第二参照直线的斜率，第四边线的斜率大于第二参照直线的斜率。例如可利用上述公式(5)计算第三边线的斜率，利用上述公式(6)计算第四边线的斜率。
在一些实施例中，在第二参照直线的斜率小于0的情况下，第三边线的斜率大于第二参照直线的斜率，第四边线的斜率小于第二参照直线的斜率。例如可利用上述公式(7)计算第三边线的斜率，利用上述公式(8)计算第四边线的斜率。
然后，变换区域生成模块1203利用第一边线、第二边线、第三边线和第四边线生成变换区域。例如，在图8的基础上，所得到的变换区域如图9中的白色线框所示。
图像处理模块1204被配置为将预设图像变换到变换区域中，以生成变换图像， 利用变换图像覆盖目标，以便更新场景图像。
在一些实施例中，图像处理模块1204通过生成变换区域的最小外接矩形，将预设图像在变换区域的最小外接矩形中重复拼接，以生成样本图像，并通过仿射变换将样本图像变换到变换区域中，以生成变换图像。
图13为本公开另一个实施例的图像处理装置的结构示意图。如图13所示，图像处理装置包括存储器1301和处理器1302。
存储器1301用于存储指令，处理器1302耦合到存储器1301，处理器1302被配置为基于存储器存储的指令执行实现如图1或图4中任一实施例涉及的方法。
如图13所示，该图像处理装置还包括通信接口1303，用于与其它设备进行信息交互。同时，该图像处理装置还包括总线1304，处理器1302、通信接口1303、以及存储器1301通过总线1304完成相互间的通信。
存储器1301可以包含高速RAM存储器，也可还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。存储器1301也可以是存储器阵列。存储器1301还可能被分块，并且块可按一定的规则组合成虚拟卷。
此外，处理器1302可以是一个中央处理器CPU，或者可以是专用集成电路ASIC，或是被配置成实施本公开实施例的一个或多个集成电路。
本公开同时还涉及一种非瞬态计算机可读存储介质，其中计算机可读存储介质存储有计算机指令，指令被处理器执行时实现如图1或图4中任一实施例涉及的方法。
在一些实施例中，在上面所描述的功能单元模块可以实现为用于执行本公开所描述功能的通用处理器、可编程逻辑控制器(Programmable Logic Controller，简称：PLC)、数字信号处理器(Digital Signal Processor，简称：DSP)、专用集成电路(Application Specific Integrated Circuit，简称：ASIC)、现场可编程门阵列(Field-Programmable Gate Array，简称：FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件或者其任意适当组合。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
本公开的描述是为了示例和描述起见而给出的，而并不是无遗漏的或者将本公开限于所公开的形式。很多修改和变化对于本领域的普通技术人员而言是显然的。选择和描述实施例是为了更好说明本公开的原理和实际应用，并且使本领域的普通技术人 员能够理解本公开从而设计适于特定用途的带有各种修改的各种实施例。

Claims (13)

  1. 一种图像处理方法，包括：
    对场景图像中的目标边界进行直线检测，以得到多个候选直线；
    在所述多个候选直线中，将与所述边界的重叠像素数量最多的候选直线作为第一参照直线；
    在所述场景图像中生成所述目标的最小外接矩形；
    生成变换区域，其中所述最小外接矩形的四个顶点位于所述变换区域的四条边线上，所述变换区域的第一边线和相对的第二边线的斜率与所述第一参照直线的斜率相关联，所述变换区域的第三边线和相对的第四边线的斜率与所述第一边线和第二边线的外角平分线的斜率相关联；
    将预设图像变换到所述变换区域中，以生成变换图像；
    利用所述变换图像覆盖所述目标，以便更新所述场景图像。
  2. 根据权利要求1所述的方法，其中，生成变换区域包括：
    生成经过所述最小外接矩形的第一顶点的第一边线和经过所述最小外接矩形的第二顶点的第二边线，所述第一顶点所在角和所述第二顶点所在角为对角，所述第一边线和所述第二边线的斜率与所述第一参照直线的斜率之差在第一预设范围内；
    生成经过所述最小外接矩形的中心点的第二参照直线，其中所述第二参照直线的斜率为所述第一边线和第二边线的外角平分线的斜率；
    生成经过所述最小外接矩形的第三顶点的第三边线和经过所述最小外接矩形的第四顶点的第四边线，所述第三顶点所在角和第四顶点所在角为对角，所述第三边线和所述第四边线的斜率与所述第二参照直线的斜率之差在第二预设范围内；
    利用所述第一边线、第二边线、第三边线和第四边线生成变换区域。
  3. 根据权利要求2所述的方法，其中，
    在所述第一参照直线的斜率小于0的情况下，所述第一边线的斜率大于所述第一参照直线的斜率，所述第二边线的斜率小于所述第一参照直线的斜率。
  4. 根据权利要求3所述的方法，其中，
    在所述第一参照直线的斜率大于0的情况下，所述第一边线的斜率小于所述第一参照直线的斜率，所述第二边线的斜率大于所述第一参照直线的斜率。
  5. 根据权利要求4所述的方法，其中，
    所述第一边线的斜率与所述第一参照直线的斜率之差为所述第一顶点到所述第一参照直线的距离和第一预设参数的比值；
    所述第二边线的斜率与所述第一参照直线的斜率之差为所述第二顶点到所述第一参照直线的距离和第二预设参数的比值。
  6. 根据权利要求5所述的方法，其中，
    在所述第二参照直线的斜率大于0的情况下，所述第三边线的斜率小于所述第二参照直线的斜率，所述第四边线的斜率大于所述第二参照直线的斜率。
  7. 根据权利要求6所述的方法，其中，
    在所述第二参照直线的斜率小于0的情况下，所述第三边线的斜率大于所述第二参照直线的斜率，所述第四边线的斜率小于所述第二参照直线的斜率。
  8. 根据权利要求7所述的方法，其中，
    所述第三边线的斜率与所述第二参照直线的斜率之差为所述第三顶点到所述第二参照直线的距离和第三预设参数的比值；
    所述第四边线的斜率与所述第二参照直线的斜率之差为所述第四顶点到所述第二参照直线的距离和第四预设参数的比值。
  9. 根据权利要求1所述的方法，其中，将预设图像映射变换到所述变换区域中，以生成变换图像包括：
    生成所述变换区域的最小外接矩形；
    将所述预设图像在所述变换区域的最小外接矩形中重复拼接，以生成样本图像；
    通过仿射变换将所述样本图像变换到所述变换区域中，以生成变换图像。
  10. 根据权利要求1-9中任一项所述的方法，还包括：
    在利用所述变换图像覆盖所述目标后，将所述变换图像中超出所述场景图像的部分删除。
  11. 一种图像处理装置，包括：
    第一处理模块，被配置为对场景图像中的目标边界进行直线检测，以得到多个候选直线，在所述多个候选直线中，将与所述边界的重叠像素数量最多的候选直线作为第一参照直线；
    第二处理模块，被配置为在所述场景图像中生成所述目标的最小外接矩形；
    变换区域生成模块，被配置为生成变换区域，其中所述最小外接矩形的四个顶点位于所述变换区域的四条边线上，所述变换区域的第一边线和相对的第二边线的斜率与所述第一参照直线的斜率相关联，所述变换区域的第三边线和相对的第四边线的斜率与所述第一边线和第二边线的外角平分线的斜率相关联；
    图像处理模块，被配置为将预设图像变换到所述变换区域中，以生成变换图像，利用所述变换图像覆盖所述目标，以便更新所述场景图像。
  12. 一种图像处理装置，包括：
    存储器，被配置为存储指令；
    处理器，耦合到存储器，处理器被配置为基于存储器存储的指令执行实现如权利要求1-10中任一项所述的方法。
  13. 一种非瞬态计算机可读存储介质，其中，非瞬态计算机可读存储介质存储有计算机指令，指令被处理器执行时实现如权利要求1-10中任一项所述的方法。
PCT/CN2022/070055 2021-02-07 2022-01-04 图像处理方法和装置 Ceased WO2022166506A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US18/263,005 US12586147B2 (en) 2021-02-07 2022-01-04 Image processing method and image processing apparatus for transforming objects within an image
EP22748785.7A EP4270316A4 (en) 2021-02-07 2022-01-04 IMAGE PROCESSING METHOD AND DEVICE

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110175113.5A CN113763439B (zh) 2021-02-07 2021-02-07 图像处理方法和装置
CN202110175113.5 2021-02-07

Publications (1)

Publication Number Publication Date
WO2022166506A1 true WO2022166506A1 (zh) 2022-08-11

Family

ID=78786607

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/070055 Ceased WO2022166506A1 (zh) 2021-02-07 2022-01-04 图像处理方法和装置

Country Status (4)

Country Link
US (1) US12586147B2 (zh)
EP (1) EP4270316A4 (zh)
CN (1) CN113763439B (zh)
WO (1) WO2022166506A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113763439B (zh) 2021-02-07 2025-03-18 北京沃东天骏信息技术有限公司 图像处理方法和装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180137654A1 (en) * 2016-11-15 2018-05-17 Fuji Xerox Co., Ltd. Image processing apparatus, image processing system, and non-transitory computer readable medium
CN109949211A (zh) * 2019-03-07 2019-06-28 北京麦哲科技有限公司 一种矩形文档图像裁切方法和装置
CN111292337A (zh) * 2020-01-21 2020-06-16 广州虎牙科技有限公司 图像背景替换方法、装置、设备及存储介质
CN112235520A (zh) * 2020-12-07 2021-01-15 腾讯科技（深圳）有限公司 一种图像处理方法、装置、电子设备及存储介质
CN113763439A (zh) * 2021-02-07 2021-12-07 北京沃东天骏信息技术有限公司 图像处理方法和装置

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10312460A (ja) * 1997-05-12 1998-11-24 Japan Aviation Electron Ind Ltd 画像処理方法及び高精度画像処理装置
JP4133246B2 (ja) * 2002-11-15 2008-08-13 日本放送協会 画像変形情報生成装置、画像変形情報生成方法及び画像変形情報生成プログラム
JP4388545B2 (ja) * 2006-12-27 2009-12-24 キヤノンソフトウェア株式会社 画像処理装置及び方法、並びにプログラム
JP5075182B2 (ja) * 2009-10-20 2012-11-14 楽天株式会社 画像処理装置、画像処理方法、および、画像処理プログラム
JP2011118554A (ja) * 2009-12-01 2011-06-16 Keyence Corp 画像処理装置、画像処理方法及びコンピュータプログラム
JP5488548B2 (ja) * 2011-08-04 2014-05-14 カシオ計算機株式会社 画像処理装置、画像処理方法及びプログラム
KR101283262B1 (ko) * 2011-10-21 2013-07-11 한양대학교 산학협력단 영상 처리 방법 및 장치
KR20140112891A (ko) * 2013-03-14 2014-09-24 삼성전자주식회사 미디어 편집을 위한 영역 선택 처리 장치, 방법 및 컴퓨터 판독 가능한 기록 매체
JP6424722B2 (ja) * 2015-04-13 2018-11-21 トヨタ自動車株式会社 四角形を補正する補正方法
US9483834B1 (en) * 2015-05-20 2016-11-01 Adobe Systems Incorporated Object boundary detection in an image
CN107240068A (zh) * 2017-05-23 2017-10-10 北京小米移动软件有限公司 图片处理方法及装置
JP2020086593A (ja) * 2018-11-16 2020-06-04 キヤノン株式会社 画像処理装置、画像処理方法
CN110738604B (zh) * 2019-07-31 2022-05-17 武汉大学 基于Canny算子和Hough变换的大倾角证件图像矫正方法及系统
CN111639642B (zh) * 2020-05-07 2023-07-07 浙江大华技术股份有限公司 一种图像处理方法、设备及装置
CN111783524B (zh) * 2020-05-19 2023-10-17 普联国际有限公司 一种场景变换检测方法、装置、存储介质及终端设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180137654A1 (en) * 2016-11-15 2018-05-17 Fuji Xerox Co., Ltd. Image processing apparatus, image processing system, and non-transitory computer readable medium
CN109949211A (zh) * 2019-03-07 2019-06-28 北京麦哲科技有限公司 一种矩形文档图像裁切方法和装置
CN111292337A (zh) * 2020-01-21 2020-06-16 广州虎牙科技有限公司 图像背景替换方法、装置、设备及存储介质
CN112235520A (zh) * 2020-12-07 2021-01-15 腾讯科技（深圳）有限公司 一种图像处理方法、装置、电子设备及存储介质
CN113763439A (zh) * 2021-02-07 2021-12-07 北京沃东天骏信息技术有限公司 图像处理方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4270316A4 *

Also Published As

Publication number Publication date
US12586147B2 (en) 2026-03-24
EP4270316A1 (en) 2023-11-01
CN113763439B (zh) 2025-03-18
EP4270316A4 (en) 2024-12-25
US20240087079A1 (en) 2024-03-14
CN113763439A (zh) 2021-12-07

Similar Documents

Publication Publication Date Title
US11961325B2 (en) Image processing method and apparatus, computer-readable medium, and electronic device
US11880977B2 (en) Interactive image matting using neural networks
CN107507217B (zh) 证件照的制作方法、装置及存储介质
US10452788B2 (en) Modeling a three-dimensional object having multiple materials
US9330466B2 (en) Methods and apparatus for 3D camera positioning using a 2D vanishing point grid
CN112712487B (zh) 一种场景视频融合方法、系统、电子设备及存储介质
CN111985036B (zh) 户型框线图绘制方法和装置、存储介质、电子设备
GB2559446A (en) Generating a three-dimensional model from a scanned object
US20190347854A1 (en) System and method for processing a graphic object
CN115439607A (zh) 一种三维重建方法、装置、电子设备及存储介质
WO2022170895A1 (zh) 图像处理方法和装置
US20170032580A1 (en) Edge preserving color smoothing of 3d models
CN113835595A (zh) 图片展示方法、装置、电子设备及计算机存储介质
CN110033507B (zh) 模型贴图的内描线绘制方法、装置、设备及可读存储介质
WO2022166506A1 (zh) 图像处理方法和装置
US10878641B1 (en) Editing bezier patch by selecting multiple anchor points
CN106934846B (zh) 一种布料图像处理方法及系统
US12400382B2 (en) Image processing method and image processing apparatus
CN111047666B (zh) 用于矢量图形和栅格图形的统一数字内容选择系统
CN119106467A (zh) 一种图像矢量化服装录入方法及系统
US11475614B2 (en) Image processing apparatus and non-transitory computer readable medium for calculating a position at which to superimpose a design object onto an image
Shen et al. Completion-based texture design using deformation
CN116647657A (zh) 响应式视频画布生成
Yen et al. Progressive completion of a panoramic image
CN114299253A (zh) 模型内景渲染方法、装置、电子设备及介质

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: 22748785

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18263005

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2022748785

Country of ref document: EP

Effective date: 20230728

NENP Non-entry into the national phase

Ref country code: DE

WWG Wipo information: grant in national office

Ref document number: 18263005

Country of ref document: US