Similar shape search system, information processing apparatus, server, computer program, and similar shape search method

Abstract

To provide a similar shape retrieval system, an information processing device, a server, a computer program, and a similar shape retrieval method that allow an accurate retrieval of a similar shape.SOLUTION: A similar shape retrieval system includes: an acquisition unit that acquires data specifying a three-dimensional shape; a feature image generation unit that generates a feature image having a pixel value indicating a three-dimensional feature at a position on the three-dimensional shape, based on the acquired data; and a retrieval unit that retrieves a similar shape that is similar to the whole or part of the three-dimensional shape, based on the generated feature image.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a similar shape search system, an information processing apparatus, a server, a computer program, and a similar shape search method.

Background Art

[0002] In recent years, CAD (Computer Aided Design) systems have been used in the design of components and the like, and the amount of accumulated CAD data has also been increasing. When designing a new component, if the accumulated CAD data is searched and utilized, past manufacturing know-how can be utilized, and the construction period can be shortened and the quality can be improved.

[0003] Non-Patent Document 1 discloses a technique for photographing a three-dimensional shape from a plurality of directions, extracting features of each image obtained by photographing, and searching for similar shapes using machine learning.

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the technique of Non-Patent Document 1, if the directions for photographing the three-dimensional shape are different, the obtained images are also different, so there is a possibility that the search results will change depending on the photographing direction. In addition, depending on the photographing direction, there is also a possibility that the features of the three-dimensional shape cannot be sufficiently captured.

[0006] The present invention has been made in view of such circumstances, and an object thereof is to provide a similar shape search system, an information processing apparatus, a server, a computer program, and a similar shape search method capable of accurately searching for similar shapes.

Means for Solving the Problems

[0007] Although this application includes a plurality of means for solving the above problems, if an example is given, the similar shape search system includes an acquisition unit that acquires data for specifying a three-dimensional shape, and based on the data acquired by the acquisition unit, a feature image generation unit that generates a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape, and a search unit that searches for a similar shape similar to all or part of the three-dimensional shape based on the feature image generated by the feature image generation unit.

Effects of the Invention

[0008] According to the present invention, similar shapes can be accurately searched for.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram showing an example of the configuration of a similar shape search system according to the present embodiment. The similar shape search system includes an information processing apparatus 10 and a server 50. The information processing apparatus 10 and the server 50 are connected via a communication network 1. The information processing apparatus 10 is a terminal device used by a user such as a designer of parts or products, and a personal computer (PC), a tablet terminal, a smartphone, or the like can be used. The server 50 may be distributed over a plurality of devices (for example, servers, PCs). When the user transmits a search processing request from the information processing apparatus 10 to the server 50, the server 50 performs a search process and transmits the search result to the information processing apparatus 10. Similar shape search means selecting, from a certain three-dimensional shape group (set), those that are determined to have a similar shape by giving all or part of a certain three-dimensional shape to the three-dimensional shape group (set).

[0011] The information processing apparatus 10 includes a control unit 11 that controls the entire apparatus, a communication unit 12, a storage unit 13, a display unit 14, an operation unit 15, and an image processing unit 16. The image processing unit 16 includes a division unit 17, a two-dimensional image generation unit 18, and a feature image generation unit 19. The control unit 11 is composed of a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.

[0012] The communication unit 12 includes a communication module and transmits and receives information (data) to and from the server 50 via the communication network 1. Also, the communication unit 12 can acquire data (for example, 3D CAD data, etc.) for specifying a three-dimensional shape from an external CAD system (not shown) or the like using wired communication or wireless communication.

[0013] The storage unit 13 can be composed of a semiconductor memory, a hard disk, or the like, and stores required information such as processing results at each part of the information processing apparatus 10 and information received from the server 50 via the communication unit 12. Also, the storage unit 13 can collect and store in advance data for specifying a three-dimensional shape.

[0014] The display unit 14 can be composed of a liquid crystal panel, an organic EL (Electro Luminescence) display, or the like.

[0015] The operation unit 15 is composed of a hardware keyboard, a mouse, or the like. Also, the operation unit 15 may be composed of a touch panel incorporated in the display unit 14. A user can perform a predetermined operation on the display unit 14. Also, the operation unit 15 may be a keyboard displayed on the display unit 14.

[0016] The image processing unit 16 has a function as an acquisition unit, acquires data for specifying a three-dimensional shape from the communication unit 12 or the storage unit 13, and generates a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape based on the acquired data.

[0017] FIG. 2 is a schematic diagram showing an example of a three-dimensional shape to be searched. As shown in FIG. 2, the user can display the three-dimensional shape to be searched on the display unit 14 by selecting three-dimensional CAD data. In the example of FIG. 2, for convenience, a relatively simple three-dimensional shape is illustrated, but actually, a more complex three-dimensional shape may be used. The user can specify all or part of the three-dimensional shape displayed on the display unit 14 as the search range. The operation unit 15 has a function as a reception unit that receives the specification of the search range among the three-dimensional shapes. The search range may be the entire three-dimensional shape or a part of the three-dimensional shape. In the example of FIG. 2, the search range is illustrated by a dashed-line frame. Also, the user can change the search range as appropriate.

[0018] The image processing unit 16 may acquire data within the specified search range among the three-dimensional shapes, and generate a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape within the search range based on the acquired data. Hereinafter, the details of the processing of the image processing unit 16 will be described.

[0019] The division unit 17 divides the three-dimensional shape into a plurality of triangular shapes based on the acquired data.

[0020] FIG. 3 is a schematic diagram showing a state of finely dividing a three-dimensional shape into a group of triangles. The curved surface of the three-dimensional shape is represented by S(u, v). (u, v) are coordinates on the curved surface S. A general method may be used for the method of finely dividing the three-dimensional shape into a group of triangles. If points on the curved surface S are S(u1, v1),..., S(un, vn),..., each vertex of the mesh of triangular shapes formed by the triangular division corresponds to each point on the curved surface S. Also, the boundary line of adjacent meshes is a straight line corresponding to a curve on the curved surface S. The size of one mesh may be determined as appropriate.

[0021] The two-dimensional image generation unit 18 generates a two-dimensional image using the plurality of triangular shapes divided by the division unit 17.

[0022] FIG. 4 is a schematic diagram showing how a triangular group is developed into a plane. A general method can be used to develop the triangular group into a plane. As shown in FIG. 4A, the triangular group is subdivided into a plurality of small pieces as shown in FIG. 4B. In the example of FIG. 4B, it is subdivided into 6 small pieces, but the number of small pieces is not limited to the example of FIG. 4B. When subdividing into small pieces, adjacent small pieces may have overlapping (common) meshes. Next, as shown in FIG. 4C, each small piece is deformed into a planar shape so that the edges (sides) of adjacent small pieces and the two vertices at both ends thereof coincide. Finally, by connecting each planar small piece in a state where the edges and both ends thereof coincide, the triangular group can be developed into a plane as shown in FIG. 4D.

[0023] FIG. 5 is a schematic diagram showing how a two-dimensional image is generated based on the result of planar development. The triangular group developed into a plane can generate a two-dimensional image Q by being arranged on a two-dimensional plane.

[0024] The feature image generation unit 19 generates a feature image by determining the pixel value of each pixel of the two-dimensional image generated by the two-dimensional image generation unit 18 so that the pixel value represents the three-dimensional feature at the position on the three-dimensional shape. That is, the feature image generation unit 19 assigns a pixel value representing a three-dimensional feature to each pixel of the two-dimensional image to generate a feature image.

[0025] FIG. 6 is a schematic diagram showing the correspondence between the pixels of the feature image P and the positions on the three-dimensional shape. The pixel (x, y) of the feature image P can be associated with the point S(u, v) of the three-dimensional shape through the corresponding position in the triangular group developed into a plane and the corresponding position in the triangular group. That is, it is possible to assign the three-dimensional feature of the position S(u, v) on the three-dimensional shape to the pixel value of the corresponding pixel (x, y) of the feature image P.

[0026] Thereby, by utilizing the property of being developable into a plane such as a press-worked part, it is possible to image as a feature image while maintaining (or almost maintaining) the information representing the three-dimensional feature of the three-dimensional shape. Therefore, the information of the three-dimensional shape is not abstracted in the process of imaging, and the information of the three-dimensional shape part can also be maintained and imaged.

[0027] In addition, since the three-dimensional shape is flat-expanded to generate the feature image, no matter what posture the three-dimensional shape is in within the three-dimensional coordinates, if it is a three-dimensional shape of the same shape, almost the same feature image can be obtained, and the feature image can be generated without depending on the posture of the three-dimensional shape.

[0028] Next, the three-dimensional features will be described.

[0029] FIG. 7 is an explanatory diagram showing an example of a method for determining the pixel value of each pixel of the feature image based on the three-dimensional features at the positions on the three-dimensional shape. The three-dimensional features may be, for example, those that represent the features of the curved surface of the three-dimensional shape. Since the three-dimensional shape is composed of a curved surface and a flat surface, the features of the curved surface may include the features of the flat surface. For parts and products such as press-processed products, manufacturing know-how is required depending on the degree of curvature of the surface, etc., which may affect the length of the construction period and the quality. Therefore, by performing a search process based on the feature image in which the features of the curved surface of the three-dimensional shape are assigned as pixel values, similar shapes with common factors such as manufacturing know-how, the length of the construction period, and the impact on quality can be searched for.

[0030] The three-dimensional features include the curvature of the three-dimensional shape. The curvature can include the maximum curvature Kmax, the minimum curvature Kmin, and the average curvature Kave {Kave = (Kmax + Kmin) / 2}. As shown in FIG. 7, the pixel value of the feature image may be the curvature K. Here, the curvature K is a function f with the maximum curvature Kmax, the minimum curvature Kmin, and the average curvature Kave as variables. K It can be calculated by. The function f K may be determined as appropriate. In this case, depending on the definition of the function f K at least one of the maximum curvature Kmax, the minimum curvature Kmin, and the average curvature Kave may be used as a variable.

[0031] Also, as shown in FIG. 7, the pixel value of the feature image may be the value of R (red), G (green), and B (blue). Here, R is a function f with the maximum curvature Kmax, the minimum curvature Kmin, and the average curvature Kave as variables. RCalculated by, and G is a function f with the maximum curvature Kmax, minimum curvature Kmin, and average curvature Kave as variables G Calculated by, and B is a function f with the maximum curvature Kmax, minimum curvature Kmin, and average curvature Kave as variables B Can be calculated by. The function f R , f G , f B May be determined as appropriate.

[0032] Also, as shown in FIG. 7, the pixel value of the feature image may be used as the value of the luminance Y. Here, the luminance Y is a function f with the maximum curvature Kmax, minimum curvature Kmin, and average curvature Kave as variables Y Can be calculated by. The function f Y May be determined as appropriate.

[0033] As described above, by using the maximum curvature Kmax and the minimum curvature Kmin as three-dimensional features, the unevenness of the three-dimensional shape can be specified. Also, by using the average curvature Kave as a three-dimensional feature, the degree of unevenness of the three-dimensional shape can be specified.

[0034] FIG. 8 is a schematic diagram showing an example of a feature image corresponding to a three-dimensional shape. As shown in FIG. 8, information such as whether the surface of the three-dimensional shape is a curved surface or a flat surface, whether it is a convex surface or a concave surface when it is a curved surface, and further the degree of curvature of the curved surface is reflected in the pixel values of the feature image. In this way, the feature image is an image of the bending characteristics of the three-dimensional shape.

[0035] The control unit 11 transmits a search processing request to the server 50 via the communication unit 12 together with the feature image generated by the image processing unit 16. Hereinafter, the server 50 will be described.

[0036] The server 50 includes a control unit 51 that controls the entire server 50, a communication unit 52, a storage unit 53, a search unit 54, a feature amount extraction unit 55, a similarity calculation unit 56, and a three-dimensional shape DB 57. The three-dimensional shape DB 57 may be provided in another data server accessible from the server 50. The control unit 51 is composed of a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.

[0037] The communication unit 52 includes a communication module and transmits and receives information (data) to and from the information processing apparatus 10 via the communication network 1. The communication unit 52 acquires (receives) a search processing request from the information processing apparatus 10 together with a feature image.

[0038] The storage unit 53 can be composed of a semiconductor memory, a hard disk, or the like, and can store a feature image received via the communication unit 52 and required information such as a processing result in each part of the server 50.

[0039] The search unit 54 searches for a similar shape that is similar to all or part of a three-dimensional shape based on the feature image generated by the information processing apparatus 10. More specifically, the search unit 54 accesses the three-dimensional shape DB 57 to search for a similar shape.

[0040] The three-dimensional shape DB 57 has a function as a storage unit and stores association information associating a feature image with a three-dimensional shape represented by the three-dimensional feature of the feature image. The three-dimensional shape DB 57 is, for example, a database in which the three-dimensional shapes of parts and products designed in the past are associated and recorded with the feature images having pixel values representing the three-dimensional features of the three-dimensional shapes.

[0041] FIG. 9 is a schematic diagram showing an example of the configuration of the three-dimensional shape DB 57. FIG. 9 shows a state in which the feature images W0001, W0002, W0003, W0004,... are stored in association with the three-dimensional shapes V0001, V0002, V0003, V0004,.... Note that the three-dimensional shape shown in FIG. 9 is a relatively simple shape for convenience, but may actually be a more complex shape. The search unit 54 compares the feature image P generated by the information processing apparatus 10 with the feature images W0001, W0002, W0003, W0004,... stored in the three-dimensional shape DB 57, and determines the similarity.

[0042] Specifically, the similarity calculation unit 56 calculates the similarity between the feature image P generated by the information processing apparatus 10 and the feature images W0001, W0002, W0003, W0004,... stored in the three-dimensional shape DB 57.

[0043] Note that, as will be described later, when calculating the feature points and feature amounts of the feature images to determine the similarity, the three-dimensional shape DB 57 may further store the feature images W0001, W0002, W0003, W0004,... in association with the feature points and feature amounts of each feature image.

[0044] FIG. 10 is a schematic diagram showing a state of comparing feature images with each other to determine the similarity. As a method of comparing feature images with each other to determine the similarity, a general method can be used. For example, the pixel values (e.g., curvature, RGB, or luminance value, etc.) are compared for all the pixels of the feature images. The similarity can be calculated by the sum of the mean squared errors of the pixel values of each pixel.

[0045] Also, as another method, the number of pixels of each pixel value (e.g., curvature, RGB, or luminance value, etc.) of the feature images is calculated, and the similarity is calculated according to the magnitude of the area of the histogram obtained by adding up the differences in the number of pixels of each feature image. It can be determined that the smaller the area, the higher the similarity.

[0046] Incidentally, the method of calculating the similarity by comparing feature images tends to increase the processing amount relatively. Therefore, for the sake of processing efficiency, feature amounts may be extracted from the feature images, and the similarity may be calculated based on the extracted feature amounts.

[0047] As shown in FIG. 10, when the size of the feature image P is smaller than the sizes of the feature images W0001, W0002, W0003, W0004, … stored in the three-dimensional shape database 57, the feature image P may be scanned over the feature images W0001, W0002, W0003, W0004, …, and the similarity with the region corresponding to the feature image P may be calculated each time of scanning, and the region with a large similarity may be determined.

[0048] FIG. 11 is a schematic diagram showing a state of determining the similarity by comparing the feature amounts of feature images. The feature amount extraction unit 55 extracts feature amounts from the feature image P generated by the information processing apparatus 10. As the feature amounts, for example, feature amounts such as SIFT (Scale Invariant Feature Transform), KAZE, AKAZE (Accelerated KAZE), SURF (Speed Up Robust Feature), and ORB (Oriented FAST and Rotated BRIEF) can be used.

[0049] In FIG. 11, the round dots indicate feature points. The extraction of the feature points can be performed, for example, by DoG (Difference-of-Gaussian) processing. Specifically, smoothed images obtained by convolving the feature image with Gaussian functions having different scales are generated, and a plurality of DoG images are generated from the differences between the generated smoothed images. Using three DoG images with different scales as a set, the adjacent pixels including the upper and lower pixels of the pixel of interest and its 26-neighborhood are compared to obtain an extreme value. The obtained extreme value can be extracted as a feature point.

[0050] Next, extract the feature amounts of the feature points. Specifically, obtain the gradient intensity and gradient direction of each pixel in the smoothed image where the feature points are detected, and set the orientation of the maximum value of the histogram quantized in 36 directions for all directions as the representative orientation of the feature points. Taking the direction of the representative orientation as the coordinate axis, divide the peripheral region of the feature points into 16 blocks of 4 blocks on each side, and generate a gradient direction histogram in 8 directions for each block. A feature amount of 128 dimensions of 4 blocks × 4 blocks × 8 directions is obtained.

[0051] To calculate the similarity based on the feature amounts, for example, obtain the distance between the feature amounts of the feature points of the feature image P and the feature amounts of the feature points of the feature image W0001 corresponding to the feature points, and sum them up for a predetermined number of feature points. As the distance of the feature amounts, for example, a norm, a Hamming distance, etc. can be used. It can be determined that the smaller the total distance, the greater the similarity.

[0052] The control unit 51 transmits the search result to the information processing apparatus 10 via the communication unit 52. The search result includes all or part of the 3D shape similar to the 3D shape corresponding to the feature image generated by the information processing apparatus 10 (3D shape). When there are a plurality of similar shapes, the similarity of each similar shape, and at least one of the display data for displaying the portions similar to the 3D shape corresponding to the feature image in different display modes.

[0053] Next, a display example of the search result on the information processing apparatus 10 will be described.

[0054] FIG. 12 is a schematic diagram showing a first example of a search result screen. In the left region of the search result screen 101, information regarding the three-dimensional shape to be searched is displayed, and in the right region, information regarding the search results is displayed. The user can select the part to be searched from the part list column 102. When the part to be searched is selected, the three-dimensional shape is displayed in the data display column 103. The user sets a required search range 104 (a dashed-line frame in FIG. 12) within the three-dimensional shape displayed in the data display column 103, and operates the "Search Range Specify" icon 105, whereby the search range of the three-dimensional shape is specified. When the user operates the "Search" icon 106, the information processing apparatus 10 generates a feature image P from the three-dimensional shape within the specified search range, and transmits a search processing request together with the feature image P to the server 50. Note that the specification of the search range is not essential. When the specification of the search range is not accepted, the entire three-dimensional shape becomes the search range.

[0055] As search results, similar shape display columns 107 and 109 are displayed for each similarity rank. Note that in the example of FIG. 12, two similar shape display columns are shown, but the number is not limited to two. In the similar shape display column 107, the most similar shape to the three-dimensional shape is displayed. Also, as indicated by the solid-line frame 108, among the similar shapes, the portions similar to the three-dimensional shape within the search range 104 can be highlighted. Note that the highlighting is an example, and can be displayed in various display modes such as changing the color, changing the pattern, or adding a mark, as long as the portions similar to the three-dimensional shape within the search range can be easily identified among the similar shapes. In the similar shape display column 109, the second most similar shape is displayed. Also, as indicated by the solid-line frame 110, among the similar shapes, the portions similar to the three-dimensional shape within the search range 104 can be highlighted. Also, as shown in FIG. 12, the similarity degree and the part name are displayed corresponding to the similar shape display columns 107 and 109.

[0056] FIG. 13 is a schematic diagram showing a second example of a search result screen. The user can appropriately change the search range among the three-dimensional shapes of the selected search target parts. This is because it is necessary to check in advance the past manufacturing know-how, quality problems, etc. across multiple locations in one three-dimensional shape. In the example of FIG. 13, a required search range 122 (a dashed-line frame in FIG. 13) is set, and by operating the "Search Range Specify" icon 105, the specification of the search range of the three-dimensional shape can be changed.

[0057] As search results on the search result screen 121, similar shape display columns 123 and 125 are displayed for each similarity rank. It can be seen that the similar shapes displayed in the similar shape display columns 123 and 125 are different. Also, as shown by the solid-line frames 124 and 126, among the similar shapes, the portions similar to the three-dimensional shapes within the search range 122 can be highlighted.

[0058] Next, the processing of the similar shape search system (similar shape search method) of the present embodiment will be described.

[0059] FIG. 14 is a flowchart showing an example of the processing procedure of the similar shape search system. The information processing apparatus 10 acquires three-dimensional shape data (S11) and accepts the specification of the search range (S12). The information processing apparatus generates a feature image from the three-dimensional shapes within the search range (S13), and transmits the generated feature image and the search processing request to the server 50 (S14).

[0060] The server 50 receives the feature image together with the search processing request (S15), calculates the feature points and feature amounts of the received feature image (S16). The server 50 searches the three-dimensional shape DB 57, calculates the similarity between the received feature image and the stored feature images, and extracts the feature images similar to the received feature image. Thereby, the server 50 can specify the similar shapes similar to the three-dimensional shape.

[0061] The server 50 transmits the three-dimensional shape or a part thereof (similar shape) determined to be similar to the information processing apparatus 10 (S18) and ends the process. When the server 50 transmits the search result to the information processing apparatus 10, the search result may include the similarity of the similar shape and display data for displaying, in different display modes, portions of the similar shape that are similar to the three-dimensional shape corresponding to the feature image.

[0062] The information processing apparatus 10 receives and displays the search result (S19) and ends the process.

[0063] As described above, the information processing apparatus 10 includes an acquisition unit that acquires data for specifying a three-dimensional shape, a feature image generation unit that generates a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape based on the acquired data, a transmission unit that transmits the generated feature image to the server, and a reception unit that receives from the server a similar shape that is similar to all or part of the three-dimensional shape.

[0064] The information processing apparatus 10 can also be realized by using a general-purpose computer including a CPU (processor), a ROM, and a RAM (memory). That is, a computer program defining the procedure of each process of the information processing apparatus 10 as shown in FIG. 14 is loaded into a RAM (memory) provided in the computer, and the information processing apparatus 10 can be realized on the computer by executing the computer program with the CPU (processor). The computer program may be recorded and distributed on a recording medium, or may be installed in the computer via a network.

[0065] The server 50 includes a storage unit that stores association information associating a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape with the three-dimensional shape, an acquisition unit that acquires a feature image corresponding to the three-dimensional shape, and a search unit that searches for a similar shape that is similar to all or part of the three-dimensional shape corresponding to the acquired feature image based on the acquired feature image.

[0066] The server 50 can also be realized by using a general-purpose computer equipped with a CPU (processor), ROM, and RAM (memory). That is, as shown in FIG. 14, a computer program that defines the procedure of each process of the server 50 is loaded into the RAM (memory) provided in the computer, and the server 50 can be realized on the computer by executing the computer program with the CPU (processor). The computer program may be recorded and distributed on a recording medium, or may be installed in the computer via a network.

[0067] In the above-described embodiment, the information processing apparatus 10 is configured to generate the feature image P, but the present invention is not limited to this, and the server 50 may be configured to generate the feature image P. In this case, the image processing unit 16 (the dividing unit 17, the two-dimensional image generation unit 18, and the feature image generation unit 19) shown in FIG. 1 may be incorporated into the server 50. The information processing apparatus 10 transmits data for specifying a three-dimensional shape for which a search range is specified to the server 50. The server 50 receives the data for specifying the three-dimensional shape and generates a feature image based on the received data.

[0068] Generally, the difficulty of searching for similar shapes of three-dimensional shapes lies in the fact that the possible postures of the three-dimensional shapes in the three-dimensional space are enormous. For example, if the three-dimensional shape is rotated one by one, it is necessary to consider 360×360 = 129,600 postures, which requires an enormous processing time. Also, when searching for similar parts from the whole of a three-dimensional shape, it is necessary to move the search area little by little for determination, which requires an enormous processing time.

[0069] According to the present embodiment, since the three-dimensional shape is two-dimensionally imaged into a feature image having pixel values representing the three-dimensional features on the three-dimensional shape and the similar shape is searched using the two-dimensionally imaged feature image, the difficulty of the above-described general search for similar shapes of three-dimensional shapes can be eliminated. Also, according to the present embodiment, since the similarity can be determined by comparing the feature images with each other, it can be realized by simple image processing.

[0070] The similar shape search system according to this embodiment includes an acquisition unit that acquires data for specifying a three-dimensional shape, and a feature image generation unit that generates a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape based on the data acquired by the acquisition unit, and a search unit that searches for a similar shape similar to all or part of the three-dimensional shape based on the feature image generated by the feature image generation unit.

[0071] The similar shape search system according to this embodiment includes a storage unit that stores association information associating a feature image with a three-dimensional shape whose three-dimensional features are represented by the feature image, and the search unit accesses the storage unit to search for a similar shape.

[0072] The similar shape search system according to this embodiment includes a division unit that divides the three-dimensional shape into a plurality of triangular shapes based on the data acquired by the acquisition unit, and a two-dimensional image generation unit that generates a two-dimensional image using the plurality of triangular shapes divided by the division unit, and the feature image generation unit generates a feature image by assigning pixel values representing the three-dimensional features to each pixel of the two-dimensional image.

[0073] In the similar shape search system according to this embodiment, the three-dimensional feature includes a feature of a curved surface of the three-dimensional shape.

[0074] In the similar shape search system according to this embodiment, the three-dimensional feature includes the curvature of the three-dimensional shape.

[0075] In the similar shape search system according to this embodiment, the three-dimensional feature includes at least one of the maximum curvature, the minimum curvature, and the average curvature of the three-dimensional shape.

[0076] The similar shape search system according to this embodiment includes a reception unit that receives a designation of a search range among the three-dimensional shapes, and the feature image generation unit generates a feature image corresponding to the search range.

[0077] The similar shape search system according to this embodiment includes a calculation unit that calculates the similarity of the similar shapes searched by the search unit, and a similarity output unit that outputs the similarity calculated by the calculation unit.

[0078] The similar shape search system according to this embodiment includes a display data output unit that outputs display data for displaying, in different display modes, portions of the similar shapes searched by the search unit that are similar to the three-dimensional shape.

[0079] The information processing apparatus according to this embodiment includes an acquisition unit that acquires data for specifying a three-dimensional shape, a feature image generation unit that generates a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape based on the data acquired by the acquisition unit, a transmission unit that transmits the feature image generated by the feature image generation unit to a server, and a reception unit that receives, from the server, a similar shape that is similar to all or part of the three-dimensional shape.

[0080] The server according to this embodiment includes a storage unit that stores association information associating a feature image having pixel values representing three-dimensional features at positions on a three-dimensional shape with the three-dimensional shape, an acquisition unit that acquires a feature image corresponding to the three-dimensional shape, and a search unit that searches, based on the feature image acquired by the acquisition unit, for a similar shape that is similar to all or part of the three-dimensional shape corresponding to the feature image.

[0081] The computer program according to this embodiment causes a computer to execute a process of acquiring data for specifying a three-dimensional shape, generating, based on the acquired data, a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape, outputting the generated feature image to a server, and acquiring, from the server, a similar shape that is similar to all or part of the three-dimensional shape.

[0082] The computer program according to this embodiment causes a computer to execute a process of acquiring a feature image corresponding to a three-dimensional shape, and searching for a similar shape similar to all or part of the three-dimensional shape corresponding to the acquired feature image, using association information associating the feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape with the three-dimensional shape.

[0083] The method for searching for a similar shape according to this embodiment includes acquiring data for specifying a three-dimensional shape, generating a feature image having pixel values representing three-dimensional features at positions on the three-dimensional shape based on the acquired data, and searching for a similar shape similar to all or part of the three-dimensional shape based on the generated feature image.

Explanation of Reference Numerals

[0084] 1 Communication network 10 Information processing apparatus 11 Control unit 12 Communication unit 13 Storage unit 14 Display unit 15 Operation unit 16 Image processing unit 17 Division unit 18 Two-dimensional image generation unit 19 Feature image generation unit 50 Server 51 Control unit 52 Communication unit 53 Storage unit 54 Search unit 55 Feature amount extraction unit 56 Similarity calculation unit 57 Three-dimensional shape DB

Claims

1. An acquisition unit that acquires data for specifying a three-dimensional shape; Based on the data acquired by the acquisition unit, the pixel value of the pixel associated with the point of the three-dimensional shape is determined through the position of each vertex of the triangular mesh of the triangular group obtained by finely dividing the three-dimensional shape, and the maximum curvature, minimum curvature, and average curvature representing the characteristics of the curved surface of the three-dimensional shape are used as variables. A feature image generation unit that generates a feature image having the pixel value including an RGB value or a luminance value calculated by a function; A search unit that searches for a similar shape similar to all or part of the three-dimensional shape based on the feature image generated by the feature image generation unit Comprising A similar shape search system.

2. Comprising a storage unit that stores association information associating a feature image with a three-dimensional shape whose three-dimensional features are represented by the feature image, The search unit Accesses the storage unit to search for a similar shape, The similar shape search system according to claim 1.

3. A division unit that divides the three-dimensional shape into a plurality of triangular shapes based on the data acquired by the acquisition unit; A two-dimensional image generation unit that unfolds the plurality of triangular shapes divided by the division unit onto a plane to generate a two-dimensional image Comprising The feature image generation unit Generates a feature image by assigning a pixel value representing the three-dimensional feature to each pixel of the two-dimensional image through the position with respect to the triangular shape, The similar shape search system according to claim 1 or claim 2.

4. The three-dimensional feature Includes the characteristics of the curved surface of the three-dimensional shape, The similar shape search system according to any one of claims 1 to 3.

5. The three-dimensional feature Includes the curvature of the three-dimensional shape, The similar shape search system according to any one of claims 1 to 4.

6. The three-dimensional feature Includes at least one of the maximum curvature, minimum curvature, and average curvature of the three-dimensional shape, The similar shape search system according to any one of claims 1 to 5.

7. Comprising a reception unit that receives a designation of a search range within the three-dimensional shape, The feature image generation unit Generates a feature image corresponding to the search range, The similar shape search system according to any one of claims 1 to 6.

8. A calculation unit that calculates the similarity of the similar shape searched by the search unit; A similarity output unit that outputs the similarity calculated by the calculation unit Comprising The similar shape search system according to any one of claims 1 to 7.

9. A display data output unit that outputs display data for displaying, in different display modes, portions of the similar shapes retrieved by the retrieval unit that are similar to the three-dimensional shape. The similar shape search system according to any one of claims 1 to 8.

10. An acquisition unit that acquires data for specifying a three-dimensional shape; Based on the data acquired by the acquisition unit, the pixel value of a pixel associated with a point on the three-dimensional shape through the position of each vertex of the triangular mesh of the triangular group obtained by finely dividing the three-dimensional shape is determined to represent the three-dimensional feature at the position on the three-dimensional shape, and a feature image generation unit that generates a feature image having the pixel value including an RGB value or a luminance value calculated by a function using the maximum curvature, minimum curvature, and average curvature representing the features of the curved surface of the three-dimensional shape as variables; A transmission unit that transmits the feature image generated by the feature image generation unit to a server; A reception unit that receives, from the server, a similar shape that is similar to all or part of the three-dimensional shape and an information processing apparatus.

11. A storage unit that stores association information associating a feature image having a pixel value including an RGB value or a luminance value calculated by a function using the maximum curvature, minimum curvature, and average curvature representing the features of the curved surface of the three-dimensional shape as variables, the pixel value of a pixel associated with a point on the three-dimensional shape through the position of each vertex of the triangular mesh of the triangular group obtained by finely dividing the three-dimensional shape being determined to represent the three-dimensional feature at the position on the three-dimensional shape, and the three-dimensional shape; An acquisition unit that acquires a feature image corresponding to a three-dimensional shape; A search unit that searches for a similar shape that is similar to all or part of the three-dimensional shape corresponding to the feature image based on the feature image acquired by the acquisition unit and a server.

12. Causing a computer to acquire data for specifying a three-dimensional shape, based on the acquired data, generate a feature image having a pixel value including an RGB value or a luminance value calculated by a function using the maximum curvature, minimum curvature, and average curvature representing the features of the curved surface of the three-dimensional shape as variables, the pixel value of a pixel associated with a point on the three-dimensional shape through the position of each vertex of the triangular mesh of the triangular group obtained by finely dividing the three-dimensional shape being determined to represent the three-dimensional feature at the position on the three-dimensional shape, output the generated feature image to a server, A computer program that causes a computer to execute a process of obtaining from the server a similar shape that is similar to all or part of the three-dimensional shape. A computer program for causing a computer to execute a process. **Claim 13** A computer is caused to obtain a feature image corresponding to a three-dimensional shape, and using association information associating the feature image having pixel values including RGB values or luminance values calculated by a function having as variables the maximum curvature, minimum curvature, and average curvature representing the features of the curved surface of the three-dimensional shape, such that the pixel values of the pixels associated with the points of the three-dimensional shape through the positions of the vertices of the triangular meshes of the group of triangles obtained by subdividing the three-dimensional shape represent the three-dimensional features at the positions on the three-dimensional shape, search for a similar shape that is similar to all or part of the three-dimensional shape corresponding to the obtained feature image. A computer program for causing a computer to execute a process. **Claim 14** An acquisition unit acquires data for specifying a three-dimensional shape, and a feature image generation unit generates a feature image having pixel values including RGB values or luminance values calculated by a function having as variables the maximum curvature, minimum curvature, and average curvature representing the features of the curved surface of the three-dimensional shape, such that the pixel values of the pixels associated with the points of the three-dimensional shape through the positions of the vertices of the triangular meshes of the group of triangles obtained by subdividing the three-dimensional shape represent the three-dimensional features at the positions on the three-dimensional shape, based on the acquired data, and a search unit searches for a similar shape that is similar to all or part of the three-dimensional shape based on the generated feature image. A method for searching for a similar shape.

Images