PROGRAM, INFORMATION PROCESSING APPARATUS, AND DECISION METHOD

Abstract

To provide a program and so on which can make determination on whether or not two images are identical to each other closer to human determination result.SOLUTION: The present invention is directed to a program for making an information apparatus functioning as a difference information generating unit for generating information of difference between a first image and a second image, a regional difference detecting unit for detecting the total amount of difference in a region by summing difference in the region of the difference information generated by the difference information generating unit, a determination unit for determining, based on the total amount of difference in the region, whether or not the first image and the second image are identical to each other.SELECTED DRAWING: Figure 14

Description

[Technical field]

[0001] The present invention relates to a program, an information processing device, and a determination method. [Background technology]

[0002] There are cases where you want to determine whether two images are the same. For example, in fields such as commercial printing that require high-quality printouts, users may want to evaluate whether printouts of the same image quality can be obtained even if the printer driver is changed due to an upgrade, etc. In addition, when compressing an image, there are cases where you want to determine whether the images can be considered the same even if the image quality is reduced by compression.

[0003] As a technique for comparing images, a technique for evaluating the user's quality of experience during video distribution is known (see, for example, Patent Document 1). Patent Document 1 discloses a method for acquiring at least one of a quality degradation index of an input transcoded video and a bit rate of the input transcoded video and a bit rate of the video before transcoding, and estimating the user's quality of experience of the transcoded video based on the acquired values. Summary of the Invention [Problem to be solved by the invention]

[0004] However, the conventional judgment method has a problem that the judgment result of whether two images are the same or not is different from the judgment result of a human being. For example, when a human judges whether two images are the same or not, even though the human wants to judge whether there is a difference in image quality when looking at them, the image will be judged to be not the same due to a minute difference detected by image processing.

[0005] Also, for example, in the case of an image such as a photograph (a complex image), there are cases where a judgment result different from a human judgment result is obtained regarding whether or not the color tones are the same.

[0006] In view of the above-mentioned problems, an object of the present invention is to provide a program or the like that can bring the judgment of whether two images are the same closer to the judgment result of a human being. [Means for solving the problem]

[0007] In view of the above problem, the present invention provides an information processing device that functions as a difference information generating unit that generates difference information between a first image and a second image, an intra-region difference detecting unit that detects an intra-region difference total amount by summing up differences within an area of ​​the difference information generated by the difference information generating unit, and a determining unit that determines whether the first image and the second image are identical or not identical based on the intra-region difference total amount, The intra-region difference detection unit calculates the intra-region difference total amount for each region while moving the region, and the determination unit determines that the first image and the second image are not identical when there is even one region where the intra-region difference total amount is equal to or greater than a threshold, and further configures the information processing device as a difference amount determination unit that, when there is a region where the intra-region difference total amount is equal to or greater than a threshold, calculates a value by dividing the intra-region difference total amount for that region by a threshold, and sets a darker color to regions where the difference is equal to or greater than the threshold as the value is larger. Provide a program to make it work. Effect of the Invention

[0008] It is possible to provide a program or the like that can bring the judgment of whether two images are the same closer to the judgment result of a human being. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating the difference between two pairs of images with different contents. [Diagram 2] FIG. 1 is a configuration diagram of an example of an information processing device or an information processing system that performs identity determination. [Diagram 3] FIG. 2 is a diagram illustrating a hardware configuration of an example of an information processing device or a server. [Figure 4] FIG. 2 is an example of a functional block diagram illustrating functions of an information processing device divided into blocks. [Diagram 5] 11A to 11C are diagrams illustrating a method for generating difference information of a CMYK image by a difference information generating unit. [Figure 6] FIG. 13 is a diagram for explaining a method for detecting continuous differences that are at least large enough to be recognized by the human eye. [Figure 7] 1A and 1B are diagrams for explaining a method for detecting differences that exist within a certain region at a rate that is greater than or equal to the rate at which the human eye can perceive them; [Figure 8]FIG. 11 is a diagram illustrating a method for generating difference information. [Figure 9] 11A and 11B are diagrams illustrating an example of difference information between a reference image and a comparative image, and an example of an arrangement pattern. [Figure 10] 11A and 11B are diagrams illustrating a method for detecting a layout pattern from the layout of differences. [Figure 11] 13 is an example of a diagram showing two images in which a differential arrangement is not detected, and differential information between the two images. [Figure 12] 1 is a diagram showing an example of two images in which a difference arrangement has been detected, and difference information between the two images; [Figure 13] 11A and 11B are diagrams illustrating a method of determining identity based on a total amount of intra-region differences. [Figure 14] FIG. 11 is a flowchart illustrating an example of a procedure in which an information processing device determines whether two images are the same based on an intra-region difference total amount. [Figure 15] FIG. 2 is a diagram showing an example of two images to be compared. [Figure 16] FIG. 16 is a diagram showing pixel-level difference information calculated from the two images in FIG. [Figure 17] FIG. 2 is a diagram showing an example of two images to be compared. [Figure 18] FIG. 13 is a diagram showing an example of an image in which the A value based on the intra-region difference total amount is reflected in red. [Figure 19] 11 is an example of an image to be compared. [Figure 20] FIG. 20 is a diagram showing an example of an image in which the A value is reflected in red based on the total amount of intra-region differences between the images in FIG. 17(a) and FIG. 19. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an information processing device and a determination method performed by the information processing device will be described as an example of an embodiment of the present invention with reference to the drawings.

[0011] <Identity judgment based on the arrangement of differences> The present applicant has devised a judgment method that focuses on the arrangement of differences in order to bring the judgment of whether two images are identical closer to the judgment result of humans. In identity judgment that focuses on the arrangement of differences, an information processing device judges that images are identical if the difference between multiple images is within a range that is not perceptible to the human eye. Therefore, when each pixel determined to be perceptible by the human eye in the difference information of two images is arranged in a manner that is perceptible to the human eye, it is regarded as a visible difference (not identical). Examples of the arrangement of differences that are perceptible to the human eye include the following. (i) The pixels judged as the difference are consecutive and large enough to be discernible by the human eye. (ii) The pixels judged to be the difference exist in a certain region at a rate that is noticeable to the human eye. (iii) The pixels determined to be different exist in a specified arrangement pattern.

[0012] <Outline of Identity Determination in the Present Embodiment> The method of judging identity by focusing on the arrangement of differences was able to bring the judgment of whether two images are identical or not closer to the judgment of humans when there was a difference in the shape of the two images. However, when it came to color, the judgment of whether two images are identical or not sometimes differed from the judgment of humans. In other words, even if there was no difference in the arrangement of the differences between the two images that humans would notice, whether humans could recognize the difference depends on the content of the images, so it was not possible to judge the identity in the same way as humans can from the arrangement of the differences alone.

[0013] The following description will be given with reference to Figure 1. Figure 1 is a diagram showing the difference between two sets of images with different contents. Figures 1(a) and (b) are the first set of images to be compared, and Figures 1(d) and (e) are the second set of images to be compared. Figures 1(c) and (f) show the respective differences.

[0014] When a human compares Figures 1(a) and (b), he or she judges there to be no difference. On the other hand, when a human compares Figures 1(d) and (e), he or she judges there to be a difference. However, as shown in Figures 1(c) and (f), the differences between the two sets of images are the same or to a similar extent. Thus, even if the arrangement of the differences in two images is the same, whether or not humans can recognize the difference depends on the content of the images. In other words, when there is an overall difference in a complex image, the difference cannot be detected in the arrangement of the differences, and the information processing device cannot judge the identity in the same way as a human. This mainly applies when there is a difference in the color tone between the two images.

[0015] Furthermore, in a method of determining identity that focuses on the arrangement of differences, the extent of the differences cannot be determined unless a human visually inspects the original image, making it difficult to prioritize when a large number of differences occur after an information processing device performs an identity determination.

[0016] In view of the above, the information processing device of this embodiment compares the total amount of differences within a region in the difference information (hereinafter referred to as the total amount of differences within a region) with a threshold value to determine whether the difference is recognizable by the human eye. In this way, the arrangement of the differences can be used to determine the difference in shape, and the total amount of differences within a region can be used to determine the difference in color, making the determination of whether two images are the same or not closer to the result of human judgment.

[0017] <Terminology> In this embodiment, "same" refers to only a difference that is not perceptible to the human eye, and "non-identical" refers to a difference that is perceptible to the human eye. "Same" may be called "similar," and "non-identical" may be called "dissimilar."

[0018] <Configuration example> Fig. 2 is a configuration diagram of an example of an information processing device 10 or an information processing system 100 that performs identity determination. In this embodiment, there is a system configuration (Fig. 2(a)) in which the information processing device 10 performs identity determination, and a system configuration (Fig. 2(b)) in which the server 30 performs identity determination. In the system configuration of Fig. 2(a), the information processing device 10 acquires two images stored in a storage unit, performs identity determination, and outputs the determination result on a display. The determination result may be sent by email or may be stored in the storage unit. The determination result may also be stored in the cloud.

[0019] The information processing device 10 may be, for example, a PC (Personal Computer), a tablet terminal, a PDA, a smartphone, or the like, as long as software runs on it.

[0020] In the information processing system 100 of Fig. 2(b), the information processing device 10 and the server 30 can communicate with each other via a network N. A server is a device that mainly performs information processing on a network and responds with processing results in response to a request received via the network.

[0021] In the system configuration of Fig. 2(b), the information processing device 10 transmits two or more images to the server 30. The server 30 performs identity determination for the images and transmits the determination result to the information processing device 10. The determination result may be transmitted by email or may be stored in the cloud.

[0022] In the configuration of FIG. 2(b), the server 30, as a so-called Web server, generates screen information (written in HTML, XML, script language, CSS (cascading style sheet), etc., and is mainly analyzed and displayed by browser software) of a portal screen that accepts image data, and provides it to the information processing device 10. A Web browser operates on the information processing device 10, receives the screen information, and displays a Web page. This Web page has an interface for registering two or more images held by the information processing device 10, and a user registers two or more images on the Web page and transmits them to the server 30.

[0023] The web page may be provided by a web application. A web application is software or a mechanism executed on a browser that operates by cooperation between a program written in a programming language (e.g. JavaScript (registered trademark)) that runs on a browser and a program on the web server side. By using the web application, the information processing device can dynamically change the web page.

[0024] <Hardware configuration example> 3 is a diagram showing a hardware configuration of the information processing device 10 or the server 30. Here, the hardware configuration of the information processing device 10 will be described.

[0025] As shown in FIG. 3, the information processing device 10 is constructed by a computer, and as shown in FIG. 3, it includes a CPU 501, a ROM 502, a RAM 503, an HD 504, an HDD (Hard Disk Drive) controller 505, a display 506, an external device connection I / F (Interface) 508, a network I / F 509, a bus line 510, a keyboard 511, a pointing device 512, a DVD-RW (Digital Versatile Disk Rewritable) drive 514, and a media I / F 516.

[0026] Among these, the CPU 501 controls the operation of the entire information processing device 10. The ROM 502 stores programs such as an IPL used to drive the CPU 501. The RAM 503 is used as a work area for the CPU 501. The HD 504 stores various data such as programs. The HDD controller 505 controls reading or writing of various data from or to the HD 504 under the control of the CPU 501. The display 506 displays various information such as a cursor, a menu, a window, a character, or an image. The external device connection I / F 508 is an interface for connecting various external devices. In this case, the external device is, for example, a USB (Universal Serial Bus) memory or a printer. The network I / F 509 is an interface for data communication using a communication network. The bus line 510 is an address bus, a data bus, or the like for electrically connecting each component such as the CPU 501 shown in FIG. 3.

[0027] Moreover, the keyboard 511 is a type of input means having a plurality of keys for inputting characters, numbers, various instructions, etc. The pointing device 512 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, etc. The DVD-RW drive 514 controls reading and writing of various data from a DVD-RW 513 as an example of a removable recording medium. The DVD-RW may be an optical storage medium such as a DVD-R. The media I / F 516 controls reading and writing (storing) of data from a recording medium 515 such as a flash memory.

[0028] <About the function> Fig. 4 is an example of a functional block diagram for explaining the functions of the information processing device 10 by dividing them into blocks. In Fig. 4, unless otherwise specified, the explanation is based on the functions of the information processing device 10, but the server 30 may have the same functions as the information processing device 10, although it is different in that it further has a communication unit.

[0029] The information processing device 10 has an image acquisition unit 11, a difference information generation unit 12, a difference information output unit 13, a difference information arrangement recognition unit 14, a judgment unit 15, a display control unit 16, an intra-region difference detection unit 17, a second judgment unit 18, and a difference amount determination unit 19. Each of these functional units of the information processing device 10 is a function or means realized by any of the components shown in Fig. 3 operating in response to an instruction from the CPU 501 in accordance with a program loaded from the HD 504 to the RAM 503.

[0030] The image acquisition unit 11 acquires (reads) two or more image data stored in the image data storage unit 21. The reason for acquiring two or more images is that it is possible to extract two images from three images for comparison, for example. The images to be acquired may be those designated by the user, or the images in the image data storage unit 21 may be acquired in order.

[0031] The difference information generating unit 12 generates difference information between two images of interest. For example, the difference information is generated by determining whether or not there are identical pixels (CMYK) at each pixel position determined by the resolution of the images. For an RGB image, the difference information generating unit 12 calculates the difference for each of RGB at each pixel position according to the resolution. In this case, the difference takes a value between -255 and 255 (when 1 pixel is 8 bits).

[0032] The difference information output unit 13 stores the difference information generated by the difference information generation unit 12 in the difference information storage unit 22. The difference information is two-dimensional information with the same resolution (size) as the image.

[0033] The difference information arrangement recognition unit 14 recognizes the arrangement of the differences. Details will be described later, but from the difference information, continuous differences, proportions within a certain area, and predetermined patterns are detected.

[0034] Based on the arrangement of the differences recognized by the difference information arrangement recognition unit 14, the determination unit 15 determines whether there is an arrangement of differences that is visible to humans (an arrangement of differences larger than a threshold value at which a difference can be determined by the human eye. The threshold value is determined by whether a difference can be determined between two images by the human eye). If there is a visible difference, it is determined that the two images are not the same. If there is no visible difference, it is determined that the two images are the same. That is, it is determined whether the continuous differences and the ratio within a certain area are greater than or less than the threshold value at which a difference can be determined by the human eye. When a predetermined pattern is detected, the determination unit 15 determines that they are not the same as they are. Note that these three criteria may be adopted by two or more ANDs or ORs.

[0035] The in-region difference detection unit 17 calculates the total in-region difference amount of the difference information generated by the difference information generation unit 12. The in-region difference detection unit 17 calculates the total in-region difference amount for the entire difference information (the entire two-dimensional image) while moving the region.

[0036] The second determination unit 18 determines whether the total in-region difference amount is greater than or equal to the threshold value. If there is at least one region in the difference information where the total in-region difference amount is greater than or equal to the threshold value, the second determination unit 18 determines that the two images are not the same.

[0037] The difference amount determination unit 19 determines how much the total in-region difference amount exceeds the threshold value. It can be said that the greater the total in-region difference amount exceeds the threshold value, the more easily noticeable it is to humans.

[0038] The display control unit 16 emphasizes the arrangement of the differences and the total in-region difference amount and displays them on the display 506.

[0039] <Generation of CMYK Difference Information> FIG. 5 is a diagram for explaining a method of generating difference information of a CMYK image by the difference information generation unit 12. In FIG. 5, a part of two images to be compared is enlarged and displayed. The difference information generation unit 12 determines whether there is a difference between the two images for each pixel position determined according to the resolution of the image (for example, in the case of an image with a resolution of 1980×1280, the vertical is equally divided into 1980 and the horizontal is equally divided into 1280 positions). In FIG. 5, the intersection of the meshes is the pixel position.

[0040] If there are pixels of the same color at the same pixel position in the two images, it is determined that there is no difference, and if there are no pixels of the same color, it is determined that there is a difference. The difference information generation unit 12 sets a flag at the pixel position where there is a difference. When the difference information is represented by an image, a dot is displayed at the pixel position with a flag. The difference information generation unit 12 prepares, for example, a table for about the number of pixel positions and records "presence or absence of a flag".

[0041] Alternatively, the difference information generation unit 12 may use a value corresponding to the magnitude of the difference as the difference information. For example, assume that there is the following difference at a certain pixel position between one image data and the other image data. A. With C and with K B. With C and with Y C. With C and nothing D. With C and with M In this case, it is considered that the differences are in the order of A < B < C < D. Therefore, the difference information generation unit 12 sets larger values in the above table in the order of A < B < C < D.

[0042] Also, the difference information generation unit 12 may calculate a feature amount by merging each pixel of an individual image with other pixels and then compare them. For example, pixels around the pixel position are weighted with a Gaussian filter or the like, and the difference between the weighted values is recorded in a table, or if there is a difference larger than a threshold value, it is recorded in the table that there is a difference (flag).

[0043] <Arrangement of differences> Subsequently, the recognition of the arrangement of differences will be described.

[0044] (i) The pixels judged to be the difference are continuous and large enough to be recognized by the human eye. This article explains:

[0045] The difference information layout recognition unit 14 sequentially scans the pixel positions to detect successive differences.

[0046] Fig. 6 is a diagram explaining a method for detecting differences that are continuous and large enough to be recognized by the human eye. In Fig. 6(a), the difference information arrangement recognition unit 14 scans the pixel positions in the horizontal direction. This makes it possible to detect differences 130 that are continuous in the horizontal direction and large enough to be recognized by the human eye.

[0047] 6B, the difference information arrangement recognition unit 14 scans the pixel positions vertically, thereby making it possible to detect any differences 131 that are vertically continuous and of a size that can be recognized by the human eye.

[0048] 6, the arrangement of rectangular differences is detected by scanning vertically and horizontally. The difference information arrangement recognition unit 14 detects the length of differences that continue vertically or horizontally. It is preferable that the difference information arrangement recognition unit 14 also scans in a diagonal direction.

[0049] The determination unit 15 then compares the length of the consecutive differences with a predetermined length (threshold value) and records that fact at the pixel positions of the consecutive differences that are longer than the predetermined length. The difference information arrangement recognition unit 14 prepares, for example, tables for the number of pixel positions and records "there is a difference" at all pixel positions of consecutive differences that are longer than the predetermined length.

[0050] This allows the display control unit 16 of the information processing device 10 to display these pixels in an emphasized manner, allowing the user to recognize the arrangement of the differences at a glance.

[0051] In this embodiment, the predetermined length (threshold) can be set to a constant value regardless of the image.

[0052] Next, (ii) The pixels judged to be different exist in a certain area at a rate that is noticeable to the human eye, In this case, the difference data arrangement recognition unit 14 moves a window surrounding the pixel position in the difference data in order, and judges whether or not there is a difference within this window that is greater than the proportion that can be recognized by the human eye. The size of this window corresponds to a fixed area.

[0053] Fig. 7 is a diagram for explaining a method for detecting differences that exist within a certain region at a rate that is more than noticeable to the human eye. Fig. 7 shows a window 140 of 4 x 4 pixels, but the size of the window 140 is just an example. The window 140 moves the pixel position at the center of the window 140 from the upper left corner to the right by one pixel at a time, and when it reaches the right end, it moves back one pixel to the left end. This movement is repeated until the pixel position at the center reaches the lower right corner.

[0054] A 4x4 pixel window 140 contains 5x5=25 pixel positions. The difference information arrangement recognition unit 14 counts how many pixel positions in the window are determined to have differences. For the sake of convenience in FIG. 7, 10 pixel positions are determined to have differences.

[0055] The judgment unit 15 judges whether or not "10 / 25>predetermined ratio" is satisfied, and if so, records that fact at all pixel positions in the window in a table equal to the number of pixel positions. This allows the display control unit 16 of the information processing device 10 to highlight and display those pixels, allowing the user to recognize the arrangement of the differences at a glance.

[0056] In this embodiment, the predetermined ratio (threshold) can be set to a constant value regardless of the image.

[0057] Next, (iii) The pixels determined to be different exist in a specified arrangement pattern; In this case, the difference information layout recognition unit 14 holds the designated layout pattern. The difference information layout recognition unit 14 performs pattern matching on the difference information with the layout pattern and judges whether or not there is a difference layout that matches the layout pattern.

[0058] Fig. 8 is a diagram for explaining a method for generating difference information. Fig. 8(a) shows an example of a reference image, and Fig. 8(b) shows an example of a comparison image. Fig. 8 shows a reference image and a comparison image that are clearly different for the purpose of explanation, but in reality, it may be difficult to see the difference between the reference image and the comparison image just by looking at them.

[0059] Fig. 9(a) shows difference information between the reference image and the comparative image in Fig. 8, and Fig. 9(b) is an enlarged view of the difference information. Fig. 9(c) shows an arrangement pattern 150. The arrangement pattern 150 is generated in advance as a pattern of differences that are easily perceptible by the human eye. The arrangement pattern 150 in Fig. 9(c) is an example, and may be a stripe, a circle, a geometric pattern, or the like.

[0060] When the difference such as the enlarged view of FIG. 9(b) matches the arrangement pattern 150, if the difference information arrangement recognition unit 14 determines that there is a difference, two images that a human being perceives as having a difference can be detected.

[0061] Fig. 10 is a diagram for explaining a method for detecting an arrangement pattern from a difference arrangement 160. For the sake of explanation, Fig. 10 shows a difference arrangement 160 having the same pattern as arrangement pattern 150. The difference information arrangement recognition unit 14 moves the arrangement pattern 150 from the upper left corner of the difference information 151 to the right by one pixel at a time, and when it reaches the right end, it moves back one pixel to the left end. The difference information arrangement recognition unit 14 repeats this movement until it reaches the lower right corner.

[0062] The difference information arrangement recognition unit 14 judges whether or not differences are recorded in the difference information at the same positions as the pixels of the arrangement pattern 150, and if differences are recorded at all pixel positions, it judges that the arrangement pattern 150 has been detected from the difference information 151. Alternatively, if differences are recorded at the positions of a certain number or more (a certain percentage or more), it judges that the arrangement pattern 150 has been detected from the difference information 151.

[0063] In FIG. 10, a difference arrangement 160 that coincides with the arrangement pattern 150 is located at the lower right of the difference information 151, and the difference information arrangement recognition unit 14 can detect this difference arrangement 160.

[0064] The difference information arrangement recognition unit 14 records, in a table for the number of pixel positions, the pixel positions that match the arrangement pattern 150 to that effect. This allows the display control unit 16 of the information processing device 10 to highlight and display these pixels, allowing the user to recognize the arrangement of the differences at a glance.

[0065] <Example of difference placement> A comparison between a case where a difference arrangement is detected and a case where it is not detected will be described with reference to FIG. 11 and FIG.

[0066] First, Fig. 11 shows two images in which no difference arrangement is detected, and the difference information between the two images. Fig. 11(a) is the reference image, Fig. 11(b) is a comparison image compressed with JPEG, and Fig. 11(c) is the difference information. Note that Fig. 11 is a CMYK image, but it can be applied to RGB as well.

[0067] When viewed by humans, there is no difference between the images in Fig. 11(a) and Fig. 11(b). However, as shown in the difference information in Fig. 11(c), it can be seen that there is an overall difference that resembles noise. Conventionally, this noise has been detected as a difference between the two images, and the information processing device 10 has sometimes determined that the images are not identical.

[0068] In this embodiment, since the information processing device 10 does not recognize the arrangement of the differences in the difference information, it can determine that the images in FIGS. 11(a) and (b) are the same.

[0069] Fig. 12 shows two images in which the arrangement of differences was detected, and the difference information between the two images. Fig. 12(a) is the reference image, Fig. 12(b) is a comparison image in which a white rectangle 170 has been added to the reference image, and Fig. 12(c) is the difference information. Note that Fig. 12 is a CMYK image, but the same can be applied to RGB.

[0070] When viewed by a human, the images in Figures 12(a) and 12(b) are clearly different. In the difference information in Figure 12(c), a rectangular difference arrangement 171 is detected at the same pixel position as the white rectangle 170 in the comparison image. In Figure 12, the difference arrangement 171 is black and white for drawing purposes, but since it is highlighted in red, for example, on the display 506, it is displayed in a manner different from the noise in Figure 11(c). Therefore, the user can determine at a glance whether or not there is a difference arrangement.

[0071] In this embodiment, since the information processing device 10 recognizes the arrangement of the differences in the difference information, it can determine that the images in FIGS. 12(a) and (b) are not the same.

[0072] <Identity determination based on the total difference in the region> 13A and 13B are diagrams for explaining a method of determining identity based on the intra-region difference total amount, in which (a) and (b) are images to be compared (one example of a first image and one example of a second image, respectively).

[0073] First, the difference information generating unit 12 smoothes Fig. 13(a) and (b). This is because if the images are left as CMYK images, each dot stands out, and a difference different from that seen by human vision is generated. That is, human vision does not compare each dot individually, but looks at the entire periphery of each dot, so the difference information generating unit 12 smoothes the CMYK images with noticeable dots. If a Gaussian filter is used for smoothing, the CMYK image is converted into an RGB image.

[0074] Figures 13(c) and 13(d) show RGB images generated by a Gaussian filter. Comparing the enlarged area 201 in Figures 13(a) and 13(c) and the enlarged area 202 in Figures 13(b) and 13(d), it can be seen that the dots have been smoothed.

[0075] The difference information generating unit 12 generates difference information for the images in Figures 13(c) and 13(d). Since the RGB images in Figures 13(c) and 13(d) have pixel values ​​of, for example, 0 to 255, when the pixel values ​​of the RGB image in Figure 13(d) are subtracted from the pixel values ​​of the RGB image in Figure 13(c), a difference in the range of -255 to 255 may occur for each pixel.

[0076] In calculating the intra-region difference total amount, negative values ​​cancel out positive values, so the difference information generating unit 12 maps differences between -255 and 255 to values ​​between 0 and 255. For example, the difference information generating unit 12 performs mapping as follows. -255 and -254 → 0 -253 and -252 → 1 : -1 and 0 → 127 1 → 128 2 and 3 → 129 : 254 and 255 → 255 Therefore, the original difference of 0 corresponds to 127, and in the mapped image, 127 means that there is no difference. FIG. 13(e) shows the mapped difference information. The difference information in FIG. 13(e) is almost gray, and there is no difference in the arrangement between the two images. Note that the difference information generating unit 12 only maps from 0 to 255 in order to fully use the 8-bit scale, but it may also map from 0 to 127 or from 0 to 511.

[0077] Then, intra-region difference detection unit 17 calculates the sum of the pixel values ​​of region 210 for each RGB. Intra-region difference detection unit 17 shifts region 210 to the right by one pixel at a time and calculates the sum in the same manner. When region 210 reaches the right end, intra-region difference detection unit 17 shifts region 210 down by one pixel to return to the left end, and performs scanning to the right.

[0078] Then, intra-region difference detection unit 17 compares the sum with a threshold for each region 210, and records the regions 210 for which a sum equal to or greater than the threshold is obtained. Thus, the coordinates of the regions 210 with differences become clear. Intra-region difference detection unit 17 records the positions of the regions 210 for which the sum equals or exceeds the threshold in each of the RGB images.

[0079] Figure 13(f) shows image data in which the areas where the sum exceeds the threshold are highlighted in black. Figure 13(f) shows a synthesis of the locations of areas 210 in each of the RGB images where the sum exceeds the threshold. This makes it easy for the user to identify areas where there are differences.

[0080] <Processing procedure for identity determination based on the total amount of differences within a region> FIG. 14 is an example of a flowchart showing a procedure in which the information processing device 10 of this embodiment determines the identity of two images based on the intra-region difference total amount.

[0081] First, the image acquisition unit 11 acquires two images to be compared from the image data storage unit 21 (S1).

[0082] The difference information generating unit 12 aligns the orientations of the two images (S2). For example, if the vertical size is H and the horizontal size is W, the difference information generating unit 12 aligns the orientations of the two images so that the aspect ratio of the two images is H:W. The orientation of either the comparison image or the reference image may be changed.

[0083] Furthermore, the difference information generating unit 12 aligns the resolution of the two images (S3). In other words, the difference information generating unit 12 aligns the sizes of the two images. The difference information generating unit 12 reduces or enlarges the comparison image to match the reference image. Steps S2 and S3 enable accurate comparison of the two images.

[0084] Next, the difference information generating unit 12 smoothes the two images using a Gaussian filter or the like (S4). This converts the CMYK image into an RGB image. An average filter or a median filter may be used for the smoothing.

[0085] The difference information generating unit 12 calculates the difference in pixel value for each RGB image (S5). If RGB is 8 bits, the difference has a value between -255 and 255.

[0086] The difference information generating unit 12 maps the differences between -255 and 255 to values ​​between 0 and 255 (S6). This mapping can be any conversion that converts negative values ​​to positive values, and the mapping destination is not limited to 0 to 255.

[0087] Next, intra-region difference detection unit 17 calculates an intra-region difference total amount by summing up the differences within the regions while moving between the regions for each RGB image (S7).

[0088] Then, the second judgment unit 18 judges whether or not there is a region where the intra-region difference total amount is equal to or greater than a threshold (S8). The second judgment unit may use a different threshold for each RGB image to judge whether or not there is a region where the intra-region difference total amount is equal to or greater than a threshold. This is because the sensitivity of human vision differs depending on RGB.

[0089] If there is even one region where the intra-region difference total amount is equal to or greater than the threshold value, the second judgment section 18 judges that the two images are not identical (S9).

[0090] If there is an area where the intra-area difference total is equal to or greater than the threshold, the difference amount determination unit 19 calculates "total value divided by threshold value" for each RGB image (S10). This calculation result is set as the "A value." A larger A value indicates a larger difference.

[0091] The difference amount determination unit 19 weights and sums the A values ​​of the regions at the same position in the RGB images (S11). The weighting is determined taking into consideration the sensitivity of the human eye to RGB.

[0092] Note that instead of summing up the A values ​​of the regions at the same position of the RGB images, only the A value of the RGB image having the largest A value in the region at the same position of the RGB images may be used. Alternatively, the difference amount determination unit 19 may use the A value of one of RGB (for example, G) in consideration of the sensitivity of the human eye.

[0093] The display control unit 16 generates an image by reflecting the A value, for example, in the density of red (S12). The display control unit 16 converts the A value into a reddish color between (0,0,0) and (255,0,0) using, for example, a predetermined table. As a result, the larger the A value, the darker the red will be. Therefore, the user can understand that the difference is greater in areas where darker red is formed.

[0094] <Comparative Example> Figures 15 and 16 are diagrams showing a comparison example of two images based on the intra-region difference sum. Figures 15(a) and (b) show two images to be compared. These images are different in the strict sense, but are judged to be indistinguishable by human vision.

[0095] Fig. 16 shows pixel-level difference information calculated from the two images in Fig. 15. As shown in Fig. 16, there are differences across the images, but no area is detected where the total amount of intra-area differences is equal to or greater than the threshold. Therefore, the information processing device 10 can output an identity judgment result similar to that of a human judgment.

[0096] 17 and 18 are diagrams showing a comparison example of two images based on the intra-region difference sum. Figures 17(a) and (b) show two images to be compared. These images are judged to have differences by human vision.

[0097] Figure 18 shows an image in which the A value based on the intra-region difference sum is reflected in red. As shown in Figure 18, areas with differences based on the intra-region difference sum are colored red, so the user can easily determine where the differences are.

[0098] Figure 19 is a comparison image different from Figure 17(b). The human judgment result shows that the difference between Figure 19 and Figure 17(a) is greater than that of Figure 17(b).

[0099] Figure 20 is an image in which the A value is reflected in red based on the total amount of intra-region differences between the images in Figures 17(a) and 19. As shown in Figure 20, areas with differences based on the total amount of intra-region differences are colored red. The red color in Figure 20 is generally darker than that in Figure 18.

[0100] In this way, since the degree of difference can be understood from the density of the red color, there is no need for a human to visually check the original image. Therefore, after the information processing device executes the identity determination, it is possible to prioritize when a large number of differences occur.

[0101] <Major Effects> As described above, the information processing device of this embodiment can determine whether or not a difference is recognizable by the human eye by comparing the total amount of differences within an area in the difference information with a threshold value. It can bring the determination of whether or not two images are the same closer to the result of human judgment with respect to color differences, which are difficult to detect by the arrangement of differences.

[0102] <Other application examples> The above describes the best mode for carrying out the present invention using examples, but the present invention is not limited to these examples in any way, and various modifications and substitutions can be made within the scope that does not deviate from the gist of the present invention.

[0103] For example, the configuration example in Fig. 4 and the like is divided according to main functions to facilitate understanding of the processing by the information processing device 10. The present invention is not limited by the manner in which the processing units are divided or the names of the processing units. The processing of the information processing device 10 can be divided into more processing units depending on the processing content. Also, it can be divided so that one processing unit includes more processes.

[0104] Each function of the above-described embodiments can be realized by one or more processing circuits. Here, the term "processing circuit" in this specification includes a processor programmed to execute each function by software, such as a processor implemented by an electronic circuit, and a device such as an ASIC (Application Specific Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), or a conventional circuit module designed to execute each function described above. [Explanation of symbols]

[0105] 10. Information processing device 30 Servers 100 Information Processing Systems [Prior art documents] [Patent documents]

[0106] [Patent Document 1] JP 2016-46685 A

Claims

1. An information processing device, a difference information generating unit that generates difference information between the first image and the second image; an intra-region difference detection unit that detects an intra-region difference total amount by summing up differences within the region of the difference information generated by the difference information generation unit; a determining unit that determines whether the first image and the second image are identical or not identical based on the intra-region difference sum, the intra-region difference detection unit calculates the intra-region difference total amount for each region while moving the region; the determination unit determines that the first image and the second image are not identical when there is at least one region in which the intra-region difference total amount is equal to or greater than a threshold; The information processing device further comprises: If there is an area where the intra-area difference total amount is equal to or greater than the threshold, a value is calculated by dividing the intra-area difference total amount of the area by the threshold; a program for causing the difference amount determining unit to function as a difference amount determining unit that sets a darker color to an area where the difference is equal to or greater than a threshold value as the value increases;

2. the difference information generating unit converts the first image and the second image, which are CMYK images, into an RGB image by smoothing the first image and the second image with a Gaussian filter; Calculating the difference information having a value between −255 and 255 for each RGB image; mapping said difference information having values ​​between −255 and 255 to said difference information having values ​​between 0 and 255; the intra-region difference detection unit calculates an intra-region difference total amount by summing up differences within an area of ​​difference information having a value of 0 to 255 for each RGB image; When there is an area where the total amount of intra-area differences is equal to or greater than a threshold value, the difference amount determination unit divides the intra-region difference total amount in the region at the same position of each RGB image by a threshold value, weights the divided values ​​by a value according to the RGB image, and sums the weighted values; The program according to claim 1 , wherein the greater the sum, the darker the color that is set for the area in which the intra-area difference sum is equal to or greater than a threshold value.

3. The difference amount determination unit does not weight the intra-region difference total amount by a threshold value and sums the weighted value by a value corresponding to the RGB image, 3. The program according to claim 2, wherein the greater the value obtained by dividing the largest intra-region difference total amount by a threshold in the regions at the same position of each RGB image, the darker the color set in the region where the intra-region difference total amount is equal to or greater than the threshold.

4. a difference information generating unit that generates difference information between the first image and the second image; an intra-region difference detection unit that detects an intra-region difference total amount by summing up differences within the region of the difference information generated by the difference information generation unit; a determination unit that determines whether the first image and the second image are identical or not identical based on the intra-region difference sum, the intra-region difference detection unit calculates the intra-region difference total amount for each region while moving the region; the determination unit determines that the first image and the second image are not identical when there is at least one region in which the intra-region difference total amount is equal to or greater than a threshold; If there is an area where the intra-area difference total amount is equal to or greater than the threshold, a value is calculated by dividing the intra-area difference total amount of the area by the threshold; a difference amount determination unit that sets a darker color to an area where the difference is equal to or greater than a threshold value as the value increases; 13. An information processing device comprising:

5. A difference information generating unit generates difference information between a first image and a second image; an intra-region difference detection unit detecting an intra-region difference total amount by summing up differences within the region of the difference information generated by the difference information generation unit; a determination unit determining whether the first image and the second image are identical or not identical based on the intra-region difference sum; a step in which the intra-region difference detection unit calculates the intra-region difference total amount for each region while moving the region; a step of determining by the determination unit that the first image and the second image are not identical when there is at least one region in which the intra-region difference total amount is equal to or greater than a threshold; a difference amount determination unit, when there is an area in which the intra-area difference total amount is equal to or greater than a threshold, calculating a value by dividing the intra-area difference total amount of the area by the threshold, and setting a darker color to the area in which the difference is equal to or greater than the threshold as the value is larger; A method for determining whether or not a particular

Images