Image processing device

Abstract

To provide an image processing device that reduces the number of missing pairs in which a pair of regions is not established.SOLUTION: The image processing device includes a control unit that performs processing for dividing first drawing data into multiple first partial regions from a first region that represents an image of one page of the first drawing data, dividing second drawing data into multiple second partial regions from a second region that represents an image of one page of the second drawing data, extracting a first feature amount of a first feature point included in each first partial region and a second feature amount of a second feature point included in each second partial region, comparing the first feature amount and the second feature amount by a feature point matching method using the first feature point and the second feature point and machine learning, identifying whether a pair of the first partial region and the second partial region is established based on the comparison result, detecting a difference between the first partial region and the second partial region when the pair is established, and allowing a movement operation of the second partial region to the first partial region where the pair is not established when the pair is not established and a ratio of the first feature amount of the first partial region where the pair is not established and the second feature amount of the second partial region is equal to or greater than a threshold value.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an image processing device. [Background technology]

[0002] A technique is known in which the similarity between the images of each page of two drawing data to be subjected to difference detection is calculated, a pair of pages for which difference detection is to be performed is specified based on the similarity, and difference detection is performed between the two drawing data for which the pair of pages has been specified. Also known is a technique in which, when an area for which difference detection is to be performed is specified within one page, difference detection is performed between the specified areas for each of the areas (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-045564 A Summary of the Invention [Problem to be solved by the invention]

[0004] When detecting differences between regions, the differences may be detected using similarities between a plurality of components that each region has, such as an average pixel value or a histogram of pixel values.

[0005] However, for example, when the positions of the regions are far apart, there is a possibility that one region may not be detected as a difference detection target relative to the other region. In this case, a pair omission occurs, in which the regions are not paired. When a pair omission occurs, there is a risk that difference detection between the regions cannot be performed.

[0006] Therefore, an object of the present invention is to provide an image processing device that reduces pair omissions, in which regions are not paired. [Means for solving the problem]

[0007] The image processing device according to the present invention includes a process of dividing first drawing data into a plurality of first partial regions from a first region representing an image of one page of the first drawing data, and dividing second drawing data into a plurality of second partial regions from a second region representing an image of one page of the second drawing data, a process of extracting a first feature amount of a first feature point possessed by each of the plurality of first partial regions and a second feature amount of a second feature point possessed by each of the plurality of second partial regions, and a process of matching the first feature amount and the second feature amount based on the first feature points, the second feature points, and a predetermined feature point matching method using machine learning. The control unit executes the following processes: a process of comparing the first feature amount with the second feature amount; a process of identifying whether or not a pair of the first partial area and the second partial area is formed based on a result of the comparison of the first feature amount with the second feature amount; a process of detecting a difference between the first partial area and the second partial area if the pair is formed; and a process of allowing an operation of moving the second partial area to the first partial area with which the pair is not formed if the pair is not formed and the ratio of the first feature amount of the first partial area with which the pair is not formed and the second feature amount of the second partial area is equal to or greater than a threshold ratio. Effect of the Invention

[0008] According to the present invention, it is possible to reduce pair omissions, in which regions are not paired with each other. [Brief description of the drawings]

[0009] [Figure 1] 2 is an example of a functional configuration of an image processing device. [Diagram 2] 10 is a flowchart illustrating an example of a process executed by a control unit. [Diagram 3] (a) is an example of first medium data. (b) is an example of second medium data. (c) is an example of a moving operation. (d) is an example of a misalignment. (e) is an example of an alignment. (f) is an example of a difference image. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] First, a description will be given of the functional configuration of the image processing device 100. As shown in Fig. 1, the image processing device 100 includes a storage unit 110, a control unit 120, an input unit 130, and an output unit 140. Note that Fig. 1 shows the main functional parts of the image processing device 100.

[0012] The storage unit 110 can be realized by a memory such as a random access memory (RAM) or a read only memory (ROM), and a storage such as a hard disk drive (HDD) or a solid state drive (SSD). The control unit 120 can be realized by a processor such as a central processing unit (CPU).

[0013] The input unit 130 can be realized by an input I / F (interface) such as a USB (Universal Serial Bus) port. The input device 11 is connected to the input unit 130. Examples of the input device 11 include a keyboard, a mouse, and a touch panel. The output unit 140 can be realized by an output I / F such as a display port. The display device 12 is connected to the output unit 140. Examples of the display device 12 include a liquid crystal display. The memory unit 110, the control unit 120, the input unit 130, and the output unit 140 are connected to each other. That is, the image processing device 100 can be realized by a computer to which various hardware is connected.

[0014] A predetermined program is temporarily stored in the storage unit 110 by the control unit 120. By executing the stored predetermined program, the control unit 120 realizes various functions of the image processing device 100 and executes various processes described later. The predetermined program may correspond to a flowchart described later.

[0015] Here, the storage unit 110 includes a first medium storage unit 111, a second medium storage unit 112, and a display drawing storage unit 113. At least one of the first medium storage unit 111, the second medium storage unit 112, and the display drawing storage unit 113 may be provided in an information processing device (not shown) different from the image processing device 100. In this case, the image processing device 100 may access the information processing device and refer to the storage contents of the first medium storage unit 111, the second medium storage unit 112, etc.

[0016] The first medium storage unit 111 stores the first medium data. The first medium data is a data file or an electronic file of an information transmission medium. The information transmission medium includes documents such as reports and instructions. In this embodiment, as shown in FIG. 3(a), the first medium data 50 is a document created on multiple pages, but may be a document created on a single page. When the first medium data 50 is a document created on multiple pages, each document includes drawing data 51, 52. The drawing data 51, 52 are, for example, design drawings of finished products or parts. The shapes of the drawing data 51, 52 are different from each other. The colors of the drawing data 51, 52 may be different from each other or the same. In this embodiment, the colors of the drawing data 51, 52 are different from each other.

[0017] The second medium storage unit 112 stores the second medium data. The second medium data is also a data file or an electronic file of an information transmission medium, like the first medium data 50. In this embodiment, as shown in FIG. 3(b), the second medium data 60 is also a document created with multiple pages, but may be a document created with a single page. The number of pages of the second medium data 60 may be the same as the number of pages of the first medium data 50, or may be different. When the second medium data 60 is a document created with multiple pages, each document includes drawing data 61, 62. The drawing data 61, 62 include, for example, design drawings of finished products and parts. The shapes of the drawing data 61, 62 are different from each other. The shapes of the drawing data 61, 62 are also different from the drawing data 51, 52. The colors of the drawing data 61, 62 may be the same as each other or different from each other. In this embodiment, the colors of the drawing data 61, 62 are the same as each other and are also the same as the color of the drawing data 51.

[0018] The above-mentioned first medium data 50 is stored in the first medium storage unit 111 based on an operation on the input device 11. The second medium data 60 is similarly stored in the second medium storage unit 112. In this embodiment, the first medium data 50 represents a comparison source for the second medium data 60, and the second medium data 60 represents a comparison destination or comparison target for the first medium data 50. Therefore, hereinafter, the drawing data 51, 52 will be referred to as base drawings 51, 52 as comparison sources, and the drawing data 61, 62 will be referred to as new drawings 61, 62 as comparison destination or comparison target.

[0019] The display drawing storage unit 113 stores an image of a drawing to be displayed on the display device 12. Although details will be described later, the display drawing storage unit 113 stores an image of a drawing in which a base drawing image corresponding to the above-mentioned base drawing 51 and a new drawing image corresponding to either one of the new drawings 61 and 62 corresponding to this base drawing image are superimposed, and a portion corresponding to the difference (i.e., a difference) is expressed in a different color. The display drawing storage unit 113 also stores an image of a drawing in which a base drawing image corresponding to the above-mentioned base drawing 52 and a new drawing image corresponding to the other of the new drawings 61 and 62 corresponding to this base drawing image are superimposed, and a portion corresponding to the difference is expressed in a different color. The base drawing image is an example of the first drawing data, and the new drawing image is an example of the second drawing data. Both the base drawing image and the new drawing image are image data.

[0020] Returning to FIG. 1, the control unit 120 includes an image generating unit 121, an image processing unit 122, and a display unit 123.

[0021] The image generating unit 121 acquires the first medium data 50 from the first medium storage unit 111, and images the first medium data 50. That is, the image generating unit 121 generates a first medium image representing the first medium data 50 from the acquired first medium data 50. If the first medium data 50 is a document created with a plurality of pages, the first medium image also includes a plurality of page images. The plurality of page images included in the first medium image include the above-mentioned base drawing image.

[0022] Similarly, the image generating unit 121 acquires the second medium data 60 from the second medium storage unit 112, and images the second medium data 60. That is, the image generating unit 121 generates a second medium image representing the second medium data 60 from the acquired second medium data 60. If the second medium data 60 is a document created with a plurality of pages, the second medium image also includes a plurality of page images. The plurality of page images included in the second medium image include the above-mentioned new drawing image.

[0023] The image processing unit 122 executes various image processes on the first medium image, the second medium image, the base drawing image, and the new drawing image. Details of the image processing unit 122 will be described later, but for example, the image processing unit 122 compares a plurality of page images included in the first medium image with a plurality of page images included in the second medium image, and identifies page images that are similar to each other. In addition, the image processing unit 122 divides an area representing a page image into a plurality of unit partial areas (hereinafter simply referred to as partial areas), and identifies partial areas that correspond to each other based on a known feature point matching method using machine learning.

[0024] For example, as shown in FIG. 3(a), the image processing unit 122 divides an area representing an entire page image included in the first medium image into a plurality of partial areas 71, 72, etc., on a page-by-page basis. Also, as shown in FIG. 3(b), the image processing unit 122 divides an area representing an entire page image included in the second medium image into a plurality of partial areas 81, 82, etc., on a page-by-page basis. The shape of the partial areas may be rectangular or may be a shape other than rectangular (e.g., circular, etc.). Also, the number of partial areas corresponding to each other may be the same or different. In addition, the image processing unit 122 monochromeizes (specifically, monochrome binarizes) each of the base drawing image and the new drawing image, and assigns different colors to each of the base drawing image and the new drawing image.

[0025] As a known feature point matching method, for example, the following documents 1 and 2 can be referred to. Reference 1: PF Alcantarilla, A. Bartoli, and AJ Davison. “KAZE features”, ECCV, (2012), pp.214?227. Reference 2: P. F, Alcantarilla, J. Nuevo, A. Bartoli, “Fast Explicit Diffusion for Accelerated Features in Nonlinear Scale Spaces”, BMVC, (2013), pp.13.1-13.11.

[0026] The display unit 123 acquires an image of the drawing stored in the display drawing storage unit 113, and displays the acquired drawing on the display device 12. As described above, the display drawing storage unit 113 stores, for example, an image of the drawing in which the parts corresponding to the differences between the base drawing image and the new drawing image are expressed in different colors. Therefore, the display unit 123 acquires and displays this drawing.

[0027] Next, the process executed by the control unit 120 will be described.

[0028] 2, the image generating unit 121 images the medium data (step S1). More specifically, the image generating unit 121 acquires the first medium data 50 from the first medium storage unit 111 and generates a first medium image. The image generating unit 121 also acquires the second medium data 60 from the second medium storage unit 112 and generates a second medium image.

[0029] When the medium data is imaged, the image processing unit 122 identifies similar pages (step S2). Specifically, the image processing unit 122 compares the multiple page images included in the first medium image with the page images included in the second medium image for each page image, and calculates the similarity between the page images. When the similarity is calculated, the image processing unit 122 identifies the page images with the highest calculated similarity as similar pages. This allows similar pages to be uniquely identified.

[0030] In this embodiment, for example, page images including the characters "FIG. 1" are identified as similar pages (see FIGS. 3(a) and (b)). Also, a page image including the characters "FIG. 2" and a page image including the characters "FIG. Z" are identified as similar pages (see FIGS. 3(a) and (b)).

[0031] When a similar page is identified, the image processing unit 122 divides one region into partial regions (step S3). That is, the image processing unit 122 divides each of the two page images identified as similar pages into a plurality of partial regions from a region corresponding to the entire page image. After region division, the image processing unit 122 executes a matching process between the two page images identified as similar pages (step S4). The matching process includes various processes such as extraction, comparison, and identification, and is a process of determining pairs of partial regions included in one page image and partial regions included in the other page image using the feature point matching method described above. Details of the matching process will be described later.

[0032] After completing the matching process, the image processing unit 122 judges whether or not a pair of partial regions has been established (step S5). If a pair has not been established (step S5: NO), the image processing unit 122 executes operation allowance control (step S6). The operation allowance control is a process that allows an operation of moving one partial region to another partial region that is not paired. In this embodiment, the image processing unit 122 executes the operation allowance control when the ratio between the feature amount of one partial region that is not paired and the feature amount of the other partial region is equal to or greater than a threshold ratio.

[0033] When the operation permission control is executed, for example, the image processing unit 122 may display a predetermined message on the display device 12. The predetermined message includes, for example, a message indicating that the operation of moving one partial region to the other partial region is permitted. It is assumed that the person in charge of checking the old and new drawings will notice this predetermined message and will perform the operation of moving the other partial region to the one partial region.

[0034] When the process of step S6 ends, the image processing unit 122 judges whether or not a move operation has been detected (step S7). As shown in Fig. 3(c), the move operation is, for example, an operation of designating (e.g., clicking or tapping) a partial area 82 on the screen of the display device 12 with a pointer (specifically, a mouse pointer) or a finger 91, and moving (e.g., dragging) the partial area 82 in the designated state. If a move operation has not been detected (step S7: NO), the image processing unit 122 ends the process as shown in Fig. 2.

[0035] On the other hand, if a movement operation is detected (step S7: YES), the image processing unit 122 calculates the amount of movement (step S8). For example, the image processing unit 122 calculates the difference between the horizontal position of the partial area 82 on the screen before and after the movement based on the movement operation of the partial area 82 as the horizontal movement amount. Also, the image processing unit 122 calculates the difference between the vertical position of the partial area 82 on the screen before and after the movement based on the movement operation of the partial area 82 as the vertical movement amount.

[0036] When the process of step S8 is completed, the image processor 122 draws the moved partial area 82 on the display device 12 based on the horizontal and vertical movement amounts (step S9), and determines whether the move operation has been confirmed (step S10). For example, until the designation of the partial area 82 is released, the image processor 122 determines that the move operation has not been confirmed (step S10: NO), and repeats the processes of steps S8 and S9. When the designation of the partial area 82 is released, the image processor 122 determines that the move operation has been confirmed (step S10: YES).

[0037] When the move operation is confirmed, because the move operation is a manual operation by the person in charge as described above, a positional deviation may occur between the base drawing 52 and the new drawing 61 included in the partial area 82, as shown in Fig. 3(d). In other words, even if the person in charge attempts to align the base drawing 52 and the new drawing 61, the accuracy of the alignment may be low because it is a manual operation.

[0038] Therefore, when the move operation is confirmed, the image processor 122 performs position adjustment (step S11). For example, as the position adjustment, the image processor 122 moves the partial region 82 in at least one of the directions of up, down, left, and right in units of one pixel, and adjusts the position of the partial region 82 to the position where the base drawing 52 and the new drawing 61 overlap most closely. As a result, the positions of the base drawing 52 and the new drawing 61 are aligned, as shown in FIG. 3(e).

[0039] When the process of step S11 is completed, the image processor 122 detects the difference between the partial regions (step S12). For example, the image processor 122 converts the base drawing 52 into a monochrome image and assigns the color green alone. Green is one of the three primary colors of light, and can be expressed as (R, G, B) = (0, 255, 0) in the color expression of brightness classified into 256 gradations in the RGB format. In the brightness classified into 256 gradations, the numerical data "255" represents the brightest brightness. This process allows the base drawing 52 to be expressed in green.

[0040] Next, the image processing unit 122 converts the new drawing 61 into a monochrome image and assigns the color red alone. Red is also one of the three primary colors of light, and can be expressed as (R, G, B) = (255, 0, 0) in the RGB color expression. This process allows the new drawing 61 to be expressed in red. Alternatively, the new drawing 61 may be assigned a green color, and the base drawing 52 may be assigned a red color.

[0041] When this process is completed, the image processor 122 superimposes the base drawing 52 and the new drawing 61 on a blue background. Blue is one of the three primary colors of light, and can be expressed as (R, G, B) = (0, 0, 255) in RGB color expression. When the base drawing 52 and the new drawing 61 are superimposed on a blue background, (R, G, B) = (255, 255, 255) is obtained, and therefore white can be expressed.

[0042] In this way, the portion of the background represented in white that contains only the green base drawing 52 can be expressed in magenta including pink because (R,G,B)=(255,255,255)-(0,255,0)=(255,0,255). Also, the portion of the background represented in white that contains only the red new drawing 61 can be expressed in cyan including light blue because (R,G,B)=(255,255,255)-(255,0,0)=(0,255,255). Also, the overlapping portion of the green base drawing 52 and the red new drawing 61 in the background represented in white can be expressed in blue because (R,G,B)=(255,255,255)-(0,255,0)-(255,0,0)=(0,0,255). In this way, the image processing unit 122 can detect the difference between partial regions.

[0043] When such processing is completed, the image processing unit 122 stores the image of the drawing in the display drawing storage unit 113. When the image of the drawing is stored in the display drawing storage unit 113, the display unit 123 displays the image of the drawing on the display device 12 (step S13), and ends the processing. As a result, as shown in Fig. 3(f), the base drawing 52 and the new drawing 61 are superimposed, and an image of the drawing partially expressed in different colors appears on the display device 12.

[0044] 3(f), the differences between the base drawing 52 and the new drawing 61 appear on the display device 12 as difference images P1 and P2 of different colors. In this case, the difference image P1, which is a part of the base drawing 52, is displayed in magenta, and the difference image P2, which is a part of the new drawing 61, is displayed in cyan. And the overlapping image P3, which is the overlapping portion of the base drawing 52 and the new drawing 61, is displayed in blue.

[0045] If a pair is established by the matching process in step S4 (step S5: YES), the image processing unit 122 performs alignment (step S14). The process of step S14 is basically the same as the process of step S11, so a detailed description will be omitted. After the process of step S14 ends, the image processing unit 122 performs the processes of steps S12 and S13, and ends the process.

[0046] In this way, the image processing device 100 according to this embodiment can detect the difference between the base drawing 52 and the new drawing 61 based on the color difference using the three primary colors of light. Therefore, the amount of data processing can be reduced compared to the case of performing data processing that does not depend on the color difference using the three primary colors of light. Also, the image of the drawing that appears on the display device 12 allows the person in charge of checking the old and new drawings to easily distinguish the difference between the drawings.

[0047] Furthermore, even if partial regions are far apart from each other and thus a pair is not formed, a partial region can be moved to another partial region. This allows the person in charge to align the positions of the partial regions by moving them, and if the alignment accuracy is low, the image processing device 100 can assist in the alignment. This can reduce the occurrence of a pair not being formed, which is a pair of partial regions not being formed.

[0048] Next, the above-mentioned matching process will be described in detail. When the process of step S3 is completed, the image processing unit 122 starts the matching process. Note that the processing order of the first comparison process and the second comparison process described later may be changed.

[0049] First, the image processing unit 122 extracts a feature amount. More specifically, the image processing unit 122 extracts a feature amount of each of a plurality of partial regions of the comparison source and a feature amount of each of a plurality of partial regions of the comparison destination as a first feature amount and a second feature amount, respectively. For example, if any of the partial regions of the comparison source has a plurality of points (e.g., pixels, or local or minute regions including a plurality of pixels as elements) A, X, etc., the image processing unit 122 extracts a first feature amount of the points A, X, etc. Also, if any of the partial regions of the comparison destination has a plurality of points B, C, Y, etc., the image processing unit 122 extracts a second feature amount of the points B, C, Y, etc.

[0050] Each of the points A, X, etc. has multiple components (e.g., lightness, saturation, brightness, average pixel value (or RGB value), variance of pixel values, histogram of pixel values, etc.) as a first feature amount. Similarly, the points B, C, Y, etc. also have multiple components as a second feature amount. Therefore, the points A and X may be called first feature points, and the points B, C, Y may be called second feature points. For example, the point X has lightness as a component x1, saturation as a component x2, ..., component x n The points A, B, C, and Y are basically the same as the point X, so detailed description will be omitted.

[0051] Next, the image processing unit 122 compares the distances as a first comparison process. More specifically, first, the image processing unit 122 calculates a first distance between any one of the first feature amounts and a second feature amount that is closest to any one of the first feature amounts. Next, the image processing unit 122 calculates a second distance between any one of the first feature amounts and a second feature amount that is next closest to any one of the first feature amounts. Then, the image processing unit 122 compares the first distance with the second distance.

[0052] For example, the image processing unit 122 calculates the distance L(A,B) between points A and B as the first distance, and calculates the distance L(A,C) between points A and C as the second distance. Here, the image processing unit 122 can calculate, for example, the distance L(X,Y) by the following calculation formula. Similarly, the image processing unit 122 can calculate the distance L(A,B) and the distance L(A,C) by this calculation formula. By the distance L(X,Y), the image processing unit 122 can determine the similarity between the feature amount of point X and the feature amount of point Y. Similarly, the image processing unit 122 can determine the similarity between the feature amount of point A and the feature amount of point B, and the similarity between the feature amount of point A and the feature amount of point C, by the distance L(A,B) and the distance L(A,C). <Calculation formula> Distance L(X,Y)={(x1-y1) 2 +(x2-y2) 2 +···+(x n -y n ) 2}} 0.5

[0053] After calculating the distance L(A,B) and the distance L(A,C), the image processing unit 122 uses the number of point pairs having unique features of the elements of the partial region to select candidates for partial regions to determine pairs (hereinafter referred to as pair candidates). Specifically, the image processing unit 122 selects partial regions with the largest number of point pairs with matching features of points within the partial region as pair candidates. The image processing unit 122 determines that the features match when the distance L between the two points is the smallest compared to the distance L between other two points.

[0054] In addition, a point pair having a unique feature is required to satisfy the following inequality, which compares, for example, the distance L(A,B) between point A and point B, which is the smallest distance candidate for a pair of point A, with the distance L(A,C) between point A and point C, which is the next smallest distance after point B, which is the pair candidate for point A. If the distance L(A,B) between point A in the comparison source partial area and point B in the comparison destination partial area is the shortest, the feature values ​​of point A and point B are very similar. Therefore, the image processing unit 122 can determine that the feature values ​​match. For the predetermined value, a positive number less than 1.0 that shortens the second distance, such as 0.5 or 0.6, can be set as a threshold value. <Inequalities> Distance L(A,B)<predetermined value x distance L(A,C)

[0055] From the above, as shown in Table 1, the image processing unit 122 can select pair candidates from a plurality of partial regions (referred to as areas in Table 1) #1 to #5 that form pairs.

[0056] [Table 1]

[0057] For example, the candidates for partial area #2 and partial area #3 of the new drawing are candidates for partial area #3 of the base drawing. Here, the number of point pairs in which the feature amounts of the points of partial area #3 of the base drawing and the points of partial area #2 of the new drawing match is 80. On the other hand, the number of point pairs in which the feature amounts of the points of partial area #3 of the base drawing and the points of partial area #3 of the new drawing match is 40. Therefore, the image processing unit 122 selects partial area #3 of the base drawing and partial area #2 of the new drawing, which have a large number of point pairs, as partial area pair candidates.

[0058] Next, the image processing unit 122 compares the distance sum with the threshold distance as a second comparison process. For example, the image processing unit 122 selects several to a dozen (for example, at least two or more, such as five, ten, or fifteen) point pairs in ascending order of distance L, and compares the distance sum, which is the sum of the distances L of the selected point pairs, with the first threshold distance. Then, the image processing unit 122 selects a pair candidate if the distance sum is less than the first threshold distance. As the first threshold distance, a numerical value of several hundred, such as a threshold distance of "300" or a threshold distance of "500", is set.

[0059] On the other hand, the image processing unit 122 may select pair candidates when the average distance obtained by dividing the sum of the distances L of the selected point pairs by the number of the selected point pairs is less than the second threshold distance. Here, when several tens of numerical values ​​such as the threshold distance "30" or the threshold distance "50" are set as the second threshold distance, the image processing unit 122 can select pair candidates of a plurality of corresponding partial areas (referred to as areas in Table 2) #1 to #5 as shown in Table 2. For example, the candidates of partial areas corresponding to partial area #3 of the base drawing are the candidates of partial area #2 and partial area #3 of the new drawing.

[0060] [Table 2]

[0061] Here, the average value of the distance L between the points in the partial area #3 of the base drawing and the points in the partial area #2 of the new drawing is "10". On the other hand, the average value of the distance L between the points in the partial area #3 of the base drawing and the points in the partial area #3 of the new drawing is "13". Therefore, the image processing unit 122 selects the partial area #3 of the base drawing and the partial area #2 of the new drawing, which have a short average distance of the distance L, as partial area pair candidates. In this way, in the second comparison process, the image processing unit 122 selects partial area pair candidates for determining pairs by using the shortness of the distance between the feature amounts of the elements of the partial areas.

[0062] When this process is completed, the image processor 122 identifies whether or not a pair of partial regions is established based on each of the selected pair candidates, and finally determines a pair of a comparison source partial region and a comparison target partial region. Therefore, as shown in Tables 1 and 2, the image processor 122 determines a pair of partial regions, for example, partial region #1 of the base drawing and partial region #1 of the new drawing, as a pair of partial regions that are established.

[0063] As shown in Table 1, the image processing unit 122 selects the partial area #5 of the base drawing and the partial area #5 of the new drawing as pair candidates, but as shown in Table 2, the partial area #5 of the base drawing and the partial area #5 of the new drawing are excluded from the selection of pair candidates. This is because the average distance of the distance L between the partial area #5 of the base drawing and the partial area #5 of the new drawing exceeds the second threshold distance. Since the positions of the partial area #5 of the base drawing and the partial area #5 of the new drawing are significantly separated, the image processing unit 122 may exclude them from the selection of pair candidates. In such a case, the image processing unit 122 may determine that they have been deleted from the base drawing and added to the new drawing.

[0064] As described above, the image processing unit 122 uses the matching process to identify or determine whether partial regions are paired, and for partial regions that are not paired, a predetermined message is displayed to prompt the person in charge to move the partial region. This makes it possible to reduce the occurrence of partial regions that are not paired. Although partial regions of drawings have been described as an example, the image processing unit 122 may also identify partial regions of characters in the same way as drawings and determine whether they are paired.

[0065] Although a preferred embodiment of the present invention has been described in detail above, the present invention is not limited to such a specific embodiment, and various modifications and variations are possible within the scope of the gist of the present invention described in the claims. [Explanation of symbols]

[0066] 50 First Media Data 51,52 Base drawing 60 Second Media Data 61,62 New drawings 71,72,81,82 partial area 100 Image processing device 120 Control section

Claims

[Claim 1] A process of dividing the first drawing data into a plurality of first partial regions from a first region representing an image of one page of the first drawing data, and dividing the second drawing data into a plurality of second partial regions from a second region representing an image of one page of the second drawing data; A process of extracting a first feature amount of a first feature point that each of the plurality of first partial regions has and a second feature amount of a second feature point that each of the plurality of second partial regions has; a process of comparing the first feature amount and the second feature amount based on the first feature points, the second feature points, and a predetermined feature point matching method using machine learning; a process of identifying whether or not a pair of the first partial region and the second partial region is established based on a comparison result between the first feature amount and the second feature amount; a process of detecting a difference between the first partial region and the second partial region when the pair is established; a process of allowing a moving operation of the second partial region to the first partial region where the pair is not established, when the pair is not established and a ratio of the first feature amount of the first partial region where the pair is not established to the second feature amount of the second partial region is equal to or greater than a threshold ratio; An image processing device having a control unit that executes the above.

Images