JPH04320578A - Device and method for storing image - Google Patents

Abstract

PURPOSE:To improve the accumulation efficiency of images and to obtain an image matched with a purpose by separating an object into a solid shape model and a mapping pattern image, individually storing both the model and image and selecting a display image out of the stored image data. CONSTITUTION:An object 12 is separated into a mapping pattern image 17 consisting of a surface pattern and a solid body 18 to which a mapping pattern is stuck. Namely the solid body 18 is converted into a solid shape model in accordance with an operator's instruction and a mapping pattern image 17 from which the influence of an illuminating condition is removed is formed by an electronic computer. The shape model and the mapping pattern are individually stored and respectively selected on occasion. Consequently the accumulation efficiency of images can be improved and an image matched with a purpose can easily be obtained.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention separates a three-dimensional shape model and a mapping pattern image when storing image data of a target object, and is useful for creating advertising images, broadcasting images, simulation images, etc. The present invention relates to an image storage device and an image storage method for use as parts.

0002

SUMMARY OF THE INVENTION In conventional image storage methods, images captured by an image input device such as a video camera are usually stored as they are.

However, in this case, the viewpoint and illumination conditions at the time of input affect the image, so if it becomes necessary to obtain an image under different conditions, it is necessary to obtain an image under the new conditions. You need to re-enter the image.

[0004] Therefore, in order to eliminate the influence of the viewpoint and illumination conditions at the time of image input, the present invention
The idea is to separate the data into a three-dimensional shape model representing the dimensional shape and a mapping pattern image consisting of the surface pattern, which is unrelated to illumination conditions, and then store each data separately.

According to the present invention, even when it is desired to obtain images from various viewpoints and illumination conditions, there is no need to accumulate a large number of images input under each condition.
It is possible to improve the efficiency of image storage, and moreover, it is possible to easily obtain images that suit the purpose.

[0006]

2. Description of the Related Art In conventional image data accumulation, as shown in FIG. , are stored as they are in the image database 36, and searched and displayed as needed to obtain an output image 37.

[0007] In this case, the viewpoint and illumination conditions of the images to be stored depend on the environment at the time of inputting the image, so in order to output images under various viewpoints and illumination conditions, it is necessary to set each condition in advance. image data must be accumulated.

[0008] Furthermore, when attempting to obtain an image under arbitrary viewpoints and illumination conditions, when displaying an image 41 that has been stored in advance, as shown in FIG. An image synthesis method has been devised that separates the three-dimensional shape model 43 and the surface mapping pattern image 44 and obtains a reconstructed image 45 under given conditions. Since the conditions vary, it is necessary to know the conditions at that time before performing image synthesis.

[0009]

According to the conventional image storage method, for example, as shown in FIG. 5, there is an image 51 of an apple illuminated from the left and an image 52 of an apple illuminated from the right. If you want to obtain such images, it is necessary to accumulate images taken under those conditions in advance.

[0010] Therefore, it is impossible to prepare in advance images corresponding to all conditions so as to always satisfy the required viewpoint and illumination conditions.

[0011] Furthermore, in an image synthesis method that attempts to obtain an image under arbitrary viewpoints and illumination conditions, it is necessary to know all the various viewpoints and illumination conditions at the time of inputting the image, and if If the conditions are unknown, it will be impossible to separate the three-dimensional shape model and the mapping pattern image and reconstruct the image under different conditions.

[0012] Therefore, an object of the present invention is to eliminate the influence of the viewpoint and illumination conditions at the time of image input, and to solve the above-mentioned problems by removing the influence of the viewpoint and illumination conditions at the time of inputting an image, and creating a mapping pattern image that is independent (irrelevant) of the illumination conditions, and An object of the present invention is to provide an image storage device and an image storage method that can create a three-dimensional model to which a pattern image is pasted and store them separately.

[0013]

[Means for Solving the Problems] An image storage device according to the present invention, when storing image data of an object photographed under a given viewpoint and illumination condition, stores a target object in three separation means for separating a three-dimensional shape model representing a dimensional shape and a mapping pattern image consisting of a surface pattern; a storage means for separately storing the three-dimensional shape model and the mapping pattern image; and selecting means for selecting an image to be displayed from among the images.

Further, the image storage method according to the present invention separates an object to be stored as image data into a three-dimensional shape model representing its three-dimensional shape and a mapping pattern image consisting of a surface pattern, and Models and mapping pattern images are stored separately, and images to be displayed are selected from among the stored image data.

[0015]

According to the present invention, first, the influence of the viewpoint of the target object whose image is input and the illumination conditions are removed, and a mapping pattern image that is independent (irrelevant) of the illumination conditions can be obtained. Furthermore, a three-dimensional shape model representing the three-dimensional shape of the target object can also be independently obtained.

Specifically, as shown in FIG. 1, in order to solve the problem of viewpoint and illumination conditions when inputting an image, a target object 12 is mapped to a mapping pattern image 17 consisting of a pattern on its surface. It is separated into a solid body 18 on which a mapping pattern is pasted. That is, the solid body 12 is converted into a three-dimensional shape model according to instructions from the operator 27 (see FIG. 2),
Furthermore, a mapping pattern image 17 from which the influence of illumination conditions has been removed is created by the electronic computer 21 (see FIG. 2).

[0017]The shape model and the mapping pattern are stored separately, and each can be selected as required.

[0018]

Embodiments Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

As shown in FIG. 1, when image data of a target object 12 is input, it is separated into a three-dimensional shape model 18 representing its three-dimensional shape and a mapping pattern image 17 consisting of a surface pattern. An example will be explained.

[0020] Regarding the target object 12, the illumination is directed from the left, and the area where the light is hitting is:
Differences can be seen in the brightness of the surface of the target object in areas that are not illuminated by light.

[0021] Therefore, if input as image data in this state, the influence of the illumination conditions at the time of input will be accumulated as it appears in the image, and it will be a very difficult task to remove this influence later. Become.

Therefore, the influence of the viewpoint and illumination conditions at the time of image input is removed, and the target object is divided into a mapping pattern image 17 that is independent (irrelevant) of the illumination conditions, and a three-dimensional shape model 18 to which the mapping pattern image is pasted. Separate into

As an example of a specific separation means, first, the target object is placed on the rotary table 13. Then, the slit projector 14 projects a band-shaped light onto the side surface of the target object. This band-shaped light is then photographed by a video camera 15. At this time, a shield 16 surrounding the target object is installed in order to eliminate the influence of illumination around the target object.

The image obtained in this way has the shape of a cut when the target object is cut vertically, and since this is the contour line that forms the three-dimensional shape of the object, the image measurement method can be used to measure the surface of the object. Coordinate values can be calculated. Also,
At this time, by determining the brightness value of the portion illuminated by light, one vertical column of the mapping pattern can be obtained. Since the band-shaped light projected by the slit projector is not affected by surrounding illumination, the brightness value on the surface of the target object does not differ depending on the position.

[0025] In this way, by repeating the above calculation while rotating the rotary table little by little, it is possible to obtain the wire frame model 18 showing the three-dimensional shape of the solid at the time when the target object has rotated once. A surface mapping pattern 17 can be developed that is independent of the surrounding lighting conditions.

[0026] The three-dimensional shape model and the mapping pattern image can be stored separately and selected as needed.

FIG. 2 shows an example of an apparatus configuration for implementing the present invention. In carrying out the present invention, image data 23 input by an image input device such as a video camera 28 is separated into a shape model and a mapping pattern using the image processing computer 21 and stored. Operator 27
While looking at the image display device 24, the user operates the operating keyboard 25 and mouse 26 to select and display the desired image.

FIG. 6 is a flowchart outlining an image storage process according to one embodiment of the present invention. As shown in this figure, first of all, a target object is placed on a rotating table (step S1).

Next, a strip of light is projected onto the side surface of the target object using the slit projector (step S2).

Next, this band-shaped light is photographed by a video camera to obtain image data for one vertical row of the target object (step S3).

Next, the rotary table is rotated once to obtain a three-dimensional model of the target object and a mapping pattern image of the surface (step S4).

Finally, the obtained image data is stored separately (step S5).

[0033]

[Effects of the Invention] By implementing the present invention, the effects of the viewpoint and lighting conditions at the time of image input are removed, and a mapping pattern image that is independent (irrelevant) of the lighting conditions and a three-dimensional shape model to which the mapping pattern image is pasted can be created. can be formed.

Furthermore, when performing image synthesis, it is possible to efficiently utilize a small amount of image data and create a new synthesized image under arbitrary lighting conditions.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment in which an image of a target object 12 is captured by a television camera 15 and separated into a shape model 18 and a mapping pattern 17 and stored.

FIG. 2 is a block diagram showing an example of a device configuration for implementing image storage according to the present invention.

FIG. 3 is a diagram illustrating a prior art image storage method that inputs images as they are;

FIG. 4 is a diagram illustrating an example of processing for separating a target object into a three-dimensional shape and a mapping pattern and reconstructing a new image.

FIG. 5 is a diagram showing the difference between an object illuminated from the left and an object illuminated from the right.

FIG. 6 is a flowchart illustrating an image storage process in one embodiment of the present invention.

[Explanation of symbols]

11 Scene with input target 12 Target object to be input 13 Turntable 14 Slit floodlight 15 Video camera 16. Shielding object 17 Mapping pattern 18 3D shape model 21 Calculator 22 Image database 23 Image data 24 Display device 25 Keyboard 26 Mouse 27 Operator 28 Video camera 31 Scene to be input 32 Input target object 33 Video camera 34 Image scanner 35 Input image 36 Image database 37 Output image 41. Accumulated images 42 Object 43 Three-dimensional shape 44 Mapping pattern 45 Reconstructed image 51 Image illuminated from the left 52 Image illuminated from the right

Claims (2)

[Claims] Claim 1: A device for accumulating image data of an object photographed under given viewpoints and illumination conditions, the device storing image data of an object photographed under given viewpoints and illumination conditions, the object being a three-dimensional model representing its three-dimensional shape, and a three-dimensional shape model representing its three-dimensional shape; a separation means for separating a mapping pattern image consisting of a pattern; a storage means for separately storing the three-dimensional shape model and the mapping pattern image; and selecting an image to be displayed from among the image data stored in the storage means. An image storage device comprising a selection means for selecting. [Claim 2] The object to be stored as image data is
The three-dimensional shape model representing the three-dimensional shape and the mapping pattern image consisting of the pattern on the surface are separated, the separated three-dimensional shape model and the image data of the mapping pattern are stored separately, and the image data of the stored image data is An image storage method characterized by selecting an image to be displayed from among images.