Information processing device, information processing device control method, and program

Abstract

To provide a technique for performing machine learning using original images while protecting personal information.SOLUTION: An information processing device includes: determination means for determining, in a first image, a concealment area in which personal information is concealed; concealment means for performing concealment processing on the first image based on the concealment area; display means for displaying a second image, which is a concealed image after the concealment processing; and learning means for performing learning using the first image.SELECTED DRAWING: Figure 2

Description

[Technical Field]

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

[0002] In recent years, research in the field of image recognition has been actively conducted, and various methods for recognizing object regions in images using convolutional neural networks (CNNs) have been proposed.

[0003] In machine learning, to obtain a machine learning model with high generalization performance, it is recommended to train using images with a variety of patterns, and the number and variety of training images affect accuracy. In particular, when it comes to human faces, there are many combinations of race, gender, age, facial expression, facial direction, lighting conditions, etc., and a large variety of images is required to improve the performance of a machine learning model that recognizes human faces.

[0004] In recent years, awareness of personal information protection has grown, making it necessary to anonymize data and difficult to collect training image data without consent. Personal information includes not only people's faces, but also documents and house nameplates that show personal names, and these images must be protected using image processing techniques such as blacking out or blurring. Such confidential images allow users to upload images without fear of their personal information being leaked. For example, online communities for sharing images for machine learning can encourage users to upload images, allowing machine learning engineers to ensure a variety of training images.

[0005] Patent Document 1 discloses a technique for generating an image of a person with hidden personal information and using the image with hidden personal information for machine learning. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2019-79357 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in Patent Document 1, machine learning is performed using personal information concealed images, and there is a possibility that learning from personal information concealed images will reduce the accuracy of the machine learning model. For example, if unnatural images with partially blurred images are used for learning, unnecessary noise may be introduced during learning, which may lead to a decrease in accuracy. Furthermore, images with concealed faces cannot be used for learning machine learning models that detect the position and size of a person's face, or machine learning models that perform face recognition to determine whether people in images are the same person, and the original images are required.

[0008] In such cases, it is necessary to use original images that have not been processed to conceal personal information for machine learning, but there is a problem in that it is difficult to perform learning using original images from the perspective of protecting personal information.

[0009] The present invention has been made in view of the above-mentioned problems, and aims to provide a technique for performing machine learning using an original image while protecting personal information. [Means for solving the problem]

[0010] To achieve the above object, an information processing device according to one aspect of the present invention comprises: a determining means for determining a concealment area for concealing personal information in the first image; a concealment unit that performs a concealment process on the first image based on the concealment area; a display means for displaying a second image, which is a concealed image that has been subjected to the concealment processing; a learning means for performing learning using the first image; The present invention is characterized by comprising: [Effects of the Invention]

[0011] According to the present invention, machine learning can be performed using an original image while protecting personal information. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram showing an example of a hardware configuration of an information processing apparatus according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an example of the software configuration of an information processing apparatus according to an embodiment. [Figure 3] 1A to 1C are diagrams illustrating the detailed configuration of an information processing device according to the second, third, and fourth embodiments. [Figure 4] 1(a) to 1(f) are diagrams for explaining the first embodiment. [Figure 5] 6A and 6B are flowcharts showing the steps of an image registration process according to the first embodiment. [Figure 6] 10 is a flowchart showing the procedure of a learning process according to an embodiment. [Figure 7] 6A and 6B are flowcharts showing the procedure of a display screen generation process according to the first embodiment. [Figure 8] FIG. 10 is a diagram for explaining a second embodiment. [Figure 9] 10(a) to 10(d) are flowcharts showing the procedure of a display screen generation process according to the second embodiment. [Figure 10] FIG. 10 is a diagram illustrating a third embodiment. [Figure 11] 10(a) and 10(b) are flowcharts showing the procedure of a display screen generation process according to the third embodiment. [Figure 12] FIG. 10 is a diagram illustrating a fourth embodiment. [Figure 13] 10(a) and 10(b) are flowcharts showing the procedure of an output map confirmation process according to a fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the claimed invention. Although multiple features are described in the embodiments, not all of these multiple features are necessarily essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same reference numerals are used to designate the same or similar components, and redundant explanations will be omitted.

[0014] (Embodiment 1: Label information is superimposed and displayed) In this embodiment, an example will be described in which a confidential image is displayed to a user who is learning, and the original image is used as the image used for learning. For example, an example will be described in which label information (face label, pupil label, mouth label, category label, gender label, orientation label, etc.) is displayed on or together with the confidential image.

[0015] <Hardware configuration> 1 is a diagram showing an example of the hardware configuration of an information processing device according to an embodiment. The information processing device 200 includes a CPU (Central Processing Unit) 100, a ROM (Read Only Memory) 110, a RAM (Random Access Memory) 120, and an HDD 130. The information processing device 200 further includes an input unit 140, an information display unit 150, and a communication unit 160.

[0016] The CPU 100 is a central processing unit that performs calculations and logical decisions for various processes. The ROM 110 stores a control program. The RAM 120 is used as a temporary storage area such as the main memory and work area of ​​the CPU 100. The HDD 130 is a hard disk for storing electronic data and programs according to this embodiment. An external storage device may also be used to perform a similar function. Here, the external storage device can be realized, for example, by a medium (recording medium) and an external storage drive for realizing access to the medium. Known examples of such media include a flexible disk (FD), CD-ROM, DVD, USB memory, MO, and flash memory. The external storage device may also be a server device connected via a network.

[0017] The input unit 140 is configured with a keyboard, touch panel, etc., and accepts input from the user. The information display unit 150 is configured with a liquid crystal display, etc., and can display various data and processing results to the user. The information processing device 200 can also communicate with other devices via the communication unit 160. Instructions from the user may be received from other devices via the communication unit 160, and processing results may be output to other devices.

[0018] <Software configuration> 2 is a diagram showing an example of the software configuration of an information processing device according to an embodiment. The information processing device 200 includes a user interface unit 210 that inputs and outputs to and from the user via the input unit 140, an internal processing unit 220 that executes internal processing of the information processing device 200, and a data management unit 230 that registers and manages input data. Note that an overview will be given here, and details will be described later.

[0019] The user interface unit 210 includes an operation unit 211, a display unit 212, and a data input unit 213. The operation unit 211 accepts mouse and keyboard operations from the user, and selects images to be registered in the data management unit 230 and images to be learned by the learning unit 223. The display unit 212 displays images to be registered in the data management unit 230 and images to be learned by the learning unit 223, allowing the user to visually check the data. The data input unit 213 accepts input data such as captured image data. For example, this is an image obtained from a photographing device such as a digital camera or a surveillance camera.

[0020] The internal processing unit 220 includes a concealment area determination unit 221 , an image conversion unit 222 , a learning unit 223 , and a label assignment unit 224 .

[0021] The concealment area determination unit 221 determines an area in which personal information is concealed for image data input from the data input unit 213. The concealment area determination may be input by a user via the operation unit 211. The image conversion unit 222 performs image conversion based on the concealment area determined by the concealment area determination unit 221, thereby concealing personal information. The learning unit 223 performs machine learning using training data held by the data management unit 230. In this embodiment, a human face detector will be described as an example of the machine learning target. The detector may be a convolutional neural network (Convolutional Neural Network) or a Vision Transformer (ViT). Alternatively, it may be a Support Vector Machine (SVM) combined with a feature extractor, or various other models. While this embodiment is not limited to the above format, the face detector will be described as a CNN in this embodiment. The labeling unit 224 assigns labels to the data input from the data input unit 213. The user may input the labels via the operation unit 211.

[0022] The data management unit 230 includes a management unit 231 and management data 232. The management unit 231 registers data input from the data input unit 213 and data processed by the internal processing unit 220. The management data 232 indicates a group of data held by the data management unit 230, such as learning data input from the data input unit 213, concealment area information determined by the concealment area determination unit 221, and a learning model generated by the learning unit 223.

[0023] <Processing Overview> In this embodiment, an example will be described in which user A registers his / her own image in the data management unit 230, and then user B performs machine learning using the image registered by user A.

[0024] 4(a)-4(f) show images of a person and a dog as example images in this embodiment. Also, FIGS. 5(a)-5(b) are flowcharts showing the steps of the image registration process according to this embodiment. FIG. 6 is a flowchart showing the steps of the learning process according to this embodiment. FIGS. 7(a)-7(b) are flowcharts showing the steps of the process of generating a screen that is displayed when confirming a learning image according to this embodiment. Hereinafter, each process in the flowchart is realized by the CPU 100 executing a control program.

[0025] <Image registration process> First, a method for user A to register an image will be described with reference to the flowcharts of Figures 5(a) and 5(b). In S501, the data input unit 213 accepts input of an image from user A. For example, user A inputs image 400 shown in Figure 4(a). In S502, the display unit 212 displays image 400 to user A.

[0026] In S503, the concealment area determination unit 221 determines a concealment area for the image 400 based on a mouse operation by user A. It is assumed that a concealment area 411 shown in image 410 in FIG. 4(b) is determined by the operation by user A. In this embodiment, an example will be described in which user A performs an operation to specify an area for the image 400 displayed on the display unit 212, but the present invention is not limited to this. For example, the concealment area determination unit 221 may automatically determine a concealment area using a trained model held by the data management unit 230. When a person's face area is set as a concealment area as shown in FIG. 4(b), a face detector can be used as the trained model.

[0027] In S504, the image conversion unit 222 performs a personal information concealment process. Here, FIG. 5(b) is a flowchart showing the detailed procedure of the personal information concealment process. In S510, the image conversion unit 222 determines a method for concealing personal information. The concealment method according to this embodiment is determined to be a filling process as specified by user A, but is not limited to this. For example, any method may be used as long as it is an image conversion process that can conceal personal information, such as a blurring process or a process of converting to a similar image.

[0028] In S511, the image conversion unit 222 performs image conversion on the image 400 using the concealment method determined in S510. For example, as shown in FIG. 4(c), a personal information concealed image 420 is generated, and it can be seen that the concealment area 421 is filled in. In this embodiment, only the concealment area is subjected to image conversion, but any range of area may be subjected to image conversion as long as it is determined based on the concealment area. For example, an area that is a predetermined multiple of the size of the concealment area may be filled in.

[0029] In S512, the display unit 212 displays the generated personal information concealment image 420. Thereafter, in S513, the display unit 212 notifies user A to confirm whether or not the personal information has been concealed by the personal information concealment image 420 (for example, by displaying a Yes / No button). If user A determines that the personal information has been concealed, the personal information concealment process ends, for example, in response to user A pressing a Yes button. On the other hand, if user A determines that the personal information has not been concealed, for example, in response to user A pressing a No button, the process returns to S510 and the concealment method is determined again. In this embodiment, it is determined that the personal information has been concealed by the personal information concealment image 420, and the personal information concealment process ends. Thereafter, the process proceeds to S505.

[0030] In S505, the labeling unit 224 labels the concealed area. In this embodiment, labeling is performed automatically, but this is not limiting. User A may manually label the concealed area using the operation unit 211, or labels may be assigned to areas other than the concealed area. FIGS. 4(d) and 4(e) show the assigned label information. A labeled image 430 shown in FIG. 4(d) shows the assigned area labels, to which a face label 431, pupil labels 432 and 433, and a mouth label 434 are assigned. Label information 440 shown in FIG. 4(e) shows the classification labels of the concealed area, to which a category label of face, a gender label of female, and a direction label of 0 degrees are assigned. Here, a label is also assigned to the concealed area 411, and a human face label, which is the same type of label as the face label 431, is assigned.

[0031] In S506, the display unit 212 displays a confirmation screen. For example, the information in FIG. 4(d) and FIG. 4(e) is displayed. Thereafter, in S507, the display unit 212 confirms with user A whether the assigned label information is correct. If the label information is correct based on the answer input from the user, the process proceeds to S508. On the other hand, if the label information is incorrect, the process returns to S505 and labels are assigned again. In this embodiment, it is assumed that all the label information is correct, and the process proceeds to S508.

[0032] In S508, the management unit 231 stores the image 400, the concealment area information indicating the concealment method and the coordinates and size of the concealment area, the personal information concealed image 420, the label information, etc. as management data 232. This completes the image registration process by user A.

[0033] <Machine learning process using original images while keeping personal information confidential> Next, with reference to the flowchart in FIG. 6, a method for learning using original images while keeping personal information confidential in the process of user B performing machine learning will be described.

[0034] In S601, the learning unit 223 selects a learning dataset based on the operation of the operation unit 211 by the user B. Here, the learning dataset is an image dataset held in advance as the management data 232. The learning dataset includes the image 400 and label information shown in FIG. 4(a).

[0035] In S602, the display unit 212 displays the learning images to confirm the images to be used for learning before executing the machine learning. Here, Fig. 7(a) shows a detailed flowchart regarding the image display when the user B selects the image 400 as the image to be displayed.

[0036] In S701, the learning unit 223 selects the image 400 selected by user B as the display image by accepting operation of the operation unit 211 by user B. In S702, the learning unit 223 acquires information about the image 400 from the management unit 231. In S703, the learning unit 223 checks whether a personal information concealment area is assigned to the image 400. If this step is Yes, the process proceeds to S704. On the other hand, if this step is No, the process proceeds to S707. In this embodiment, since a personal information concealment area is assigned to the image 400, the process proceeds to S704.

[0037] In S704, the display unit 212 displays on the display screen the personal information concealed image 420 acquired from the management unit 231. Note that, although the present embodiment shows an example in which the personal information concealed image 420 is acquired from the management unit 231, the present invention is not limited to this. For example, the personal information concealed image 420 may not be stored in the management unit 231, and may be generated using the image 400 and the concealed area information.

[0038] In S705, the learning unit 223 determines whether label information has been assigned to the image 400. If this step is Yes, the process proceeds to S706. On the other hand, if this step is No, the process ends. In this embodiment, since label information has been assigned, the process proceeds to S706. In S706, the learning unit 223 adds a label list to the display screen. Details of the process of S706 will be described later with reference to FIG. 7(b). In S707, the display unit 212 displays the image 400, which is the original image, since there is no personal information concealment area. This completes the series of processes in FIG. 7(a).

[0039] Next, the detailed processing of S706 will be described with reference to the flowchart of FIG.

[0040] In S710, the learning unit 223 selects one piece of label information from the assigned label information based on the operation of the operation unit 211 by the user B. In this embodiment, it is assumed that the face label 431 is selected first.

[0041] In S711, the learning unit 223 determines whether the selected label information is an area label. If the result of this step is Yes, the learning unit 223 proceeds to S712. On the other hand, if the result of this step is No, the learning unit 223 proceeds to S716. Since the face label 431 selected in S710 is a label indicating an area in the image and includes coordinate information, the learning unit 223 proceeds to S712.

[0042] In S712, the learning unit 223 uses the coordinate information to superimpose a label on the personal information concealed image 420. In S713, the learning unit 223 adds label information to the region label list. Here, the region label list is a list of label information superimposed on the personal information concealed image 420, and is information that is displayed on the screen when the display screen is updated in S715. A detailed description will be given later. Here, the label information of the face label 431 is added to the region label list.

[0043] In S714, the learning unit 223 determines whether all labels assigned to the image 400 have been processed. If this step is Yes, proceed to S715. On the other hand, if this step is No, return to S710. Here, since there are still labels that have not been processed, return to S710 and continue processing.

[0044] In S710, the learning unit 223 next selects a category label based on the operation of the operation unit 211 by user B. In this case, in the determination process of S711, the category label is a classification label assigned to the hidden area, and the selected label does not include coordinate information, so the process proceeds to S716.

[0045] In S716, the learning unit 223 adds category label information to the classification label list. Here, the classification label list is a list of label information assigned to the hidden area, and is information that is displayed on the screen when the display screen is updated in S715. A detailed explanation will be given later. Thereafter, the process proceeds to S714, where it is determined whether all labels assigned to the image 400 have been processed. Here, since there is label information that has not been processed, the process returns to S710, and the same process continues.

[0046] By processing the assigned labels in order as described above, the region labels are superimposed on the personal information concealed image 420 and added to the region label list, and the classification labels are added to the classification label list. After processing of all labels is completed, the process proceeds to S715.

[0047] In S715, the display unit 212 updates the display screen. Here, FIG. 4(f) shows an example of a display screen 450 displayed to user B after all processing in this embodiment is completed. Region labels are superimposed on the display image 451, and information indicating the region labels is listed in the region label list 452. In the illustrated example, the dashed-dotted line indicates the face region, the solid line indicates the pupil region, and the dotted line indicates the mouth region. This allows visual confirmation of which labels are assigned to which positions, even in the case of a personal information concealment image in which the original image cannot be viewed for personal information protection purposes. Furthermore, the classification label list 453 lists classification label information for the concealment region. In the illustrated example, it can be recognized that the category is face, the gender is female, and the orientation is 0 degrees (front). This allows user B, who cannot view the original image, to know what kind of image it is. When the processing of S715 is completed, the processing of S706 is completed, and the processing of S602 is also completed.

[0048] 6, the learning unit 223 determines the learning data based on the operation of the operation unit 211 by the user B. In S604, the learning unit 223 executes machine learning. This completes the processing in FIG. 6.

[0049] As described above, in this embodiment, image data is managed by the management unit 231, the image 400 is used for machine learning, and the personal information concealed image 420 is used when displaying to the user B who is learning.

[0050] This allows machine learning to be performed using the original image while keeping personal information secret. Also, by superimposing the label on the personal information concealed image, User B, who cannot see the original image, can visually recognize the information in the original image, which helps User B decide whether or not to use the personal information concealed image for learning.

[0051] (Embodiment 2: Display of Similar Images) In this embodiment, an example will be described in which a hidden image is displayed to a user who is learning, and the image used for learning is the original image. More specifically, an example will be described in which a similar image (i.e., a modified hidden image) that is similar to the hidden area is displayed.

[0052] In this embodiment, as in embodiment 1, an example is used in which user A registers image 400 shown in Figure 4(a), which is an image of a person and a dog, in the data management unit 230, and an example is described in which an image similar to the secret area 411 shown in Figure 4(b) is displayed on the display screen.

[0053] The hardware configuration of the information processing device 200 according to this embodiment is the same as the configuration shown in FIG. 1 of the first embodiment, and the software configuration is also the same as that of the first embodiment in terms of the user interface unit 210 and the data management unit 230, but the configuration of the internal processing unit is different.

[0054] 3(a) shows an example of an internal processing unit 300 according to this embodiment. In this embodiment, the internal processing unit 300 further includes a similar image search unit 301 for displaying similar images, in addition to the components shown in embodiment 1. The similar image search unit 301 includes an image similarity calculation unit 302, a label similarity calculation unit 303, and a similar image generation unit 304.

[0055] In this embodiment, an example is described in which, after the processing of Figures 7(a) and 7(b) described in embodiment 1 is completed and the display screen 450 shown in Figure 4(f) is displayed, user B checks the images to be used for learning by displaying similar images.

[0056] FIG. 8 is a diagram showing an example of a display screen when similar images according to this embodiment are displayed. Display screen 800 is an example of a display screen after the processing according to this embodiment has been performed. A search method selection box 801 allows selection of whether or not to display similar images, and a similar image search method when similar images are displayed. In the example shown, the options listed are "Do not display similar images," "Display images with similar images," "Display images with similar labels," and "Generate similar images." First, the processing when "Display similar images" is selected in search method selection box 801 will be described.

[0057] <Display processing of similar images with high image similarity> 9A and 9B are flowcharts showing the procedure of a process for searching for a similar image according to this embodiment.

[0058] In S901, the similar image searching unit 301 determines a search method based on an input from user B. Here, user B has selected "display similar images" in the search method selection box 801. Therefore, it is determined that images with high image similarity (above a predetermined value) will be searched for using the image similarity calculation unit 302.

[0059] Here, image similarity is calculated as the similarity between images by quantifying the brightness and color distribution in the image, object position information, edge information, etc. In this embodiment, feature amounts of the images are extracted using a trained model held by the data management unit 230, and similarity comparison is performed, but this is not limited to this. For example, a method of directly comparing images may also be used.

[0060] In S902, the similar image search unit 301 determines whether or not to display similar images. If this step is Yes, the process proceeds to S903. On the other hand, if this step is No, the process proceeds to S905. For example, if "Do not display similar images" is selected in the search method selection box 801, the process proceeds to S905, and similar images are not displayed. Here, it is assumed that "Display similar images" has been selected, and the process proceeds to S903.

[0061] In S903, the similar image searching unit 301 searches for a similar image. Here, the flowchart in Fig. 9(b) shows the detailed processing of S903. In S911, the similar image searching unit 301 acquires an original image. In this case, the image 400 shown in Fig. 4(a) is acquired.

[0062] In S912, the similar image search unit 301 acquires, from the management data 232 held by the data management unit 230, images to which a person's face label has been assigned and which do not have a personal information concealment area, as candidates for images to be displayed as similar images. Here, as in the first embodiment, a person's face label has been assigned to the concealment area 411 (FIG. 4(b)). Therefore, it can be seen that an image with a person's face label is suitable for display as a similar image. Furthermore, in S912, in order to acquire candidates for images to be displayed as similar images, images that can be displayed to user B and do not have a personal information concealment area are acquired.

[0063] In S913, the image similarity calculation unit 302 calculates the image similarity. For example, the image similarity is calculated between the concealed area 411 shown in FIG. 4(b) and the face area of ​​each person in the image group acquired in S912. Note that in this embodiment, the image similarity is calculated for the concealed area, but this is not limiting. For example, the image similarity of any area may be used, such as the image similarity of the entire image or the image similarity of an area determined based on the concealed area.

[0064] In S914, the similar image search unit 301 acquires an image having a high similarity (a predetermined value or more) calculated by the image similarity calculation unit 302. This completes the processing in Fig. 9(b). After that, the process proceeds to S904.

[0065] In S904, the display unit 212 displays a similar image. For example, an image having a facial area of ​​a person similar to the hidden area of ​​the image 400 shown in FIG. 4(a) is displayed as a similar image in the similar image display unit 802 (FIG. 8). In S905, the display unit 212 does not display a similar image. This completes the series of processes in FIG. 9(a).

[0066] <Displaying similar images with high label similarity> Next, a process will be described when "display items with similar labels" is selected in the search method selection box 801 in FIG.

[0067] In S901, if user B selects "Display images with similar labels" in the search method selection box 801 in Figure 8, it is decided to search for images with high label similarity using the label similarity calculation unit 303.

[0068] Here, label similarity is a numerical representation of the degree to which label information in images matches between images, and the more the labels match, the higher the similarity. In this embodiment, the similarity is calculated so that the more the number of matching labels is, the higher the similarity, but this is not limited to this. For example, the similarity may be calculated by weighting each type of label.

[0069] In S902, it is determined whether or not to display similar images. In this case, "display images with similar labels" is selected, so the process proceeds to S903. Here, with reference to the flowchart in FIG. 9(c), the detailed process procedure of S903 when "display images with similar labels" is selected in the search method selection box 801 in FIG. 8 will be described.

[0070] In S921, the label similarity calculation unit 303 acquires the label of the original image. Here, the label of the image 400 in FIG. 4(a), which is the original image, is acquired. In S922, the label similarity calculation unit 303 acquires candidate images to be displayed as similar images from the management data 232 held by the data management unit 230. Here, images to which a human face label has been assigned and which do not have a personal information concealment area are acquired.

[0071] In S923, the label similarity calculation unit 303 calculates the label similarity. Here, the label similarity between the concealed area 411 shown in FIG. 4(b) and each human face area in the image group acquired in S922 is calculated. Here, the label similarity within the concealed area is calculated, but this is not limiting. For example, the label similarity of any area may be used, such as the label similarity of the entire image or the label similarity of an area determined based on the concealed area.

[0072] In S924, the similar image search unit 301 acquires images with high label similarity (greater than or equal to a predetermined value) based on the label similarity calculated by the label similarity calculation unit 303. This completes the processing of Fig. 9(c). Then, the process proceeds to S904.

[0073] In S904, the display unit 212 displays similar images. For example, an image having a face region of a person whose label is similar to that of the hidden region of the image 400 shown in Fig. 4(a) is displayed as a similar image in the similar image display unit 802 (Fig. 8).

[0074] <Generation and display of similar images> Finally, the process when "generate similar images" is selected in the search method selection box 801 will be described.

[0075] In S901, if user B selects "Generate similar image" in the search method selection box 801 in Fig. 8, it is determined that an image with a high degree of similarity is generated using the similar image generation unit 304. In addition, in this embodiment, generation of a similar image refers to generation of a similar image by, for example, generative artificial intelligence.

[0076] In S902, it is determined whether or not to display similar images. In this case, "Generate similar images" is selected, so the process proceeds to S903. Here, with reference to the flowchart in FIG. 9(d), the detailed process procedure of S903 when "Generate similar images" is selected in the search method selection box 801 in FIG. 8 will be described.

[0077] In S931, the similar image generation unit 304 converts the hidden area into a similar image. Here, the hidden area 411 shown in FIG. 4(b) is converted into a similar image. Note that, although an example of converting a hidden area into a similar image will be described here, the conversion into a similar image is not limited to this. For example, any area may be converted into a similar image, such as the entire image area or an area determined based on the hidden area.

[0078] In S932, the image similarity calculation unit 302 determines whether the similarity between the generated similar image and the original image is greater than or equal to a first threshold and less than or equal to a second threshold. Focusing on the original purpose of concealing personal information, the lower the similarity between the generated similar image and the original image, the better the concealment of personal information. However, if a dissimilar image is displayed as a similar image, user B may mistakenly recognize the original image as a different image. Therefore, in this embodiment, the first threshold is set to 0.2 and the second threshold is set to 0.8. By determining whether the similarity between the generated similar image and the original image is greater than or equal to the first threshold and less than or equal to the second threshold, the similar image is determined to be not too similar. Here, if the similarity is too low (similarity<first threshold) or too high (similarity>second threshold), the process of S931 is performed again. If an image with a similarity greater than or equal to the first threshold and less than or equal to the second threshold is generated, the process proceeds to S933.

[0079] In S933, the similar image search unit 301 acquires the image generated by the similar image generation unit 304. This completes the processing of Fig. 9(d). After that, the process proceeds to S904.

[0080] In S904, the display unit 212 displays the similar image. For example, the generated image is displayed as a similar image in the similar image display area 802 (FIG. 8).

[0081] As described above, in this embodiment, it is possible to display an image similar to the hidden area, thereby providing visual support to user B who is unable to know the original image. It is also possible to support the user in determining whether or not to use the personal information concealed image for learning. Therefore, user convenience is improved.

[0082] (Embodiment 3: Display of recommendation level for learning use) In this embodiment, an example will be described in which a confidential image is displayed to a user who is studying, and the original image is used as the image for studying. Also, an example will be described in which the study use information and the study use recommendation score are displayed at that time.

[0083] In this embodiment, as in embodiment 1, an example is used in which user A registers image 400 shown in Figure 4(a), which is an image of a person and a dog, in the data management unit 230, and an example is described in which the learning use information and learning use recommendation score of image 400 are displayed on the display screen.

[0084] Here, the learning use information is information associated with the original image, and can include at least one of the accuracy improvement rate for each task when other users use the image for learning, the number of times it has been used for learning, the number of likes, etc. The number of likes is the number of users who, among those who used the target image for learning, evaluated that using the image improved their learning. The learning use recommendation score is a score calculated according to formula (1), and the higher the score, the more recommended the image is for learning.

[0085]

number

[0086] where S f is the learning recommendation score for a person's face, and R f is the average improvement rate of human face detection accuracy when other users used image 400 for learning. f is the number of times other users have used image 400 for face detection training, and G f is the number of likes for the image 400. Rf , W Uf , W Gf indicates the respective weighting coefficients, and the weighting coefficients can be determined for each image and each learning task. Rf =0.02, W Uf =0.00005, W Gf =0.0002 for the explanation.

[0087] An example of the hardware configuration of the information processing device 200 according to this embodiment is the same as the configuration shown in FIG. 1 of the first embodiment, and the software configuration is also the same as that of the first embodiment in terms of the user interface unit 210 and the data management unit 230, but the configuration of the internal processing unit is different.

[0088] 3(b) shows an example of the internal processing unit 310 according to this embodiment. In this embodiment, the internal processing unit 310 further includes a recommendation level calculation unit 311 for calculating a recommendation level for use in learning, in addition to the components shown in the first embodiment.

[0089] In this embodiment, an example is described in which, after the processing of Figures 7(a) and 7(b) described in embodiment 1 is completed and the display screen 450 shown in Figure 4(f) is displayed, user B displays the learning use information and the learning use recommendation level and checks the quality of the image to be used for learning.

[0090] FIG. 10 is a diagram showing an example of a display screen when displaying learning use information and learning use recommendation level according to this embodiment. Display screen 1000 is an example of a display screen after the processing according to this embodiment is performed. A learning use information display selection box 1001 allows the user to select whether or not to display learning use information. In this embodiment, the processing when "Display learning use information" is selected in the learning use information display selection box 1001 will be described.

[0091] <Display processing of learning use information> 11(a) and 11(b) are flowcharts showing the processing steps for displaying learning use information and learning use recommendation level according to this embodiment. In S1101, the internal processing unit 310 acquires information indicating whether or not to display learning use information based on input from user B. In S1102, the internal processing unit 310 determines whether or not to display learning use information based on the information acquired in S1101. If this step is Yes, proceed to S1103. On the other hand, if this step is No, proceed to S1106. Specifically, if "Do not display learning use information" is selected in the learning use information display selection box 1001, proceed to processing in S1106, and the learning use information is not displayed. In this case, since "Display learning use information" is selected, proceed to processing in S1103.

[0092] In S1103, the internal processing unit 310 determines whether a label has been assigned to the image (for example, image 400). If the answer to this step is Yes, the process proceeds to S1104. On the other hand, if the answer to this step is No, the process proceeds to S1106. In this case, a label has been assigned to the image 400, so the process proceeds to S1104.

[0093] In S1104, the recommendation level calculation unit 311 calculates a recommendation level score for use in learning. FIG. 11B is a flowchart showing the details of the process in S1104. In S1111, the recommendation level calculation unit 311 selects one of the assigned labels. In this embodiment, the face label 431 is selected first. In S1112, the recommendation level calculation unit 311 calculates a recommendation level score for use in learning. For example, the recommendation level calculation unit 311 calculates the recommendation level score for use in learning of the face label using the learning use information 1002 in FIG. 10 (accuracy improvement rate=10, number of uses for learning=2000, number of likes=500) and formula (1). In this example, the result is 10×0.02+2000×0.00005+500×0.0002=0.4. In S1113, the recommendation level calculation unit 311 determines whether all labels assigned to the image (e.g., image 400) have been processed. If this step is Yes, the process ends. On the other hand, if this step is No, the process returns to S1111. Here, since there are labels that have not been processed, the process returns to S1111 and continues.

[0094] As described above, by processing the assigned labels in order, it is possible to calculate the scores of the recommendation levels for learning use for all labels. After processing all labels has been completed, the process proceeds to S1105.

[0095] In S1105, the display unit 212 displays the learning use information and the learning use recommendation score. In the learning use information 1002 of FIG. 10, as an example, the learning use information (accuracy improvement rate, number of times used for learning, number of likes) and the learning use recommendation score for the face label, pupil label, and gender label are listed. For example, in FIG. 10, scores including 0.4 calculated for the face label are displayed. In S1106, the display unit 212 does not display the learning use information and the learning use recommendation score. This completes the processing of FIG. 11(a).

[0096] As described above, in this embodiment, it is possible to display the learning use information and the learning use recommendation score, thereby supporting user B who is unable to know the original image. It is also possible to support the user in determining whether or not to use the personal information concealed image for learning. Therefore, user convenience is improved.

[0097] (Embodiment 4: Display of learning status) According to the above-described first to third embodiments, it is possible to perform machine learning using an original image while concealing personal information, but it is difficult to check the learning status of machine learning using an original image while concealing personal information. In contrast, this embodiment describes a method that enables personal information to be kept confidential even when checking the learning status of machine learning using an original image.

[0098] The hardware configuration of the information processing device 200 according to this embodiment is the same as the configuration shown in FIG. 1 of the first embodiment, and the software configuration is also the same as that of the first embodiment in terms of the data management unit 230, but the configurations of the internal processing unit and the user interface unit are different.

[0099] 3(c) shows an example of the internal processing unit 320 and the user interface unit 330 according to this embodiment. The internal processing unit 320 further includes a similar image search unit 301 and an output map conversion unit 321 in addition to the components of the first embodiment.

[0100] The similar image search unit 301 includes an image similarity calculation unit 302. The output map conversion unit 321 includes a model output map conversion unit 322 and a model output map conversion parameter determination unit 323. The user interface unit 330 further includes a model output map display unit 331 in addition to the components of the first embodiment.

[0101] In this embodiment, an example will be described in which, after the determination of learning data in S603 in FIG. 6 is completed and machine learning is performed in S604, an output map obtained during learning is displayed while personal information is kept confidential.

[0102] FIG. 12 shows images for explaining a method for checking a learning status according to this embodiment. Image 1200 shows a learning image used in this embodiment. The image shows a person and a dog, with the person wearing a nameplate with their name on it around their neck. Labeled image 1210 shows assigned region labels, which are a face label 1211, pupil labels 1212 and 1213, a mouth label 1214, and a nameplate label 1215. Personal information concealed image 1220 is an image obtained by performing personal information concealment processing on image 1200 registered by user A. The regions assigned with face label 1211 and nameplate label 1215 are filled in as shown in concealed regions 1221 and 1222, thereby concealing the personal information.

[0103] <Learning status confirmation process> 13(a) and 13(b) are flowcharts showing the procedure of a process for checking a learning status according to this embodiment. In S1301, the internal processing unit 320 selects a learning image. For example, the image 1200 is selected as the learning image. In S1302, the learning unit 223 performs machine learning. In this embodiment, the tasks to be learned are a person's face detection task and a person's eye detection task. Therefore, machine learning is performed using a face label 1211 and eye labels 1212 and 1213.

[0104] In S1303, the learning unit 223 acquires the output map after learning has ended. At this time, the output map is not yet displayed to user B. The maps to be output are, for example, pupil output map 1230, face output map 1240, and model intermediate output map 1250. Here, the pupil output map and face output map are maps output from the final layer of the model, and are output in the same size as the learning image. Furthermore, the model intermediate output map is a map output from the intermediate layer of the model, and is output in the same size as the learning image.

[0105] In this embodiment, an example in which these three maps are output will be described, but the present invention is not limited to this. For example, other maps such as a mouth map, a head map, and a category classification map may also be output.

[0106] It can be seen that pupil output map 1230 shows a response at the positions of both pupils, and face output map 1240 shows a response at the center of the face. By displaying such output maps, it is possible to check whether learning is progressing correctly for each task to be learned.

[0107] To more accurately confirm the accuracy of the learning, one method is to overlay the output map on the image used for learning and closely observe where on the image the map responds. However, since image 1200 is a confidential image, user B cannot overlay the output map on image 1200 to check the map's response. Note that if the elements that make up the model include a pooling layer, an upsampling layer, etc., the output map may be output at a size different from that of the learning image. In this case, it is recommended to resize either the map or the image so that they are equal in size and then overlay them.

[0108] It can also be seen that many edge features of image 1200 remain in model intermediate output map 1250. If such an output map is displayed to user B, user B will know what the original image was like, which could lead to the leakage of personal information.

[0109] From these, it can be seen that the following problems arise when trying to check the progress of learning personal information concealed images: (1) The original image cannot be used for the image on which the output map is superimposed. (2) There is a possibility that personal information may remain in the output map. Therefore, in this embodiment, support is provided so that user B can accurately know the reaction of the output map even during learning of personal information concealed images.

[0110] In S1304, the learning unit 223 selects one of the output maps. In this embodiment, the pupil output map 1230 is selected first. In S1305, the learning unit 223 determines whether or not the image 1200 includes a personal information concealment area in the selected output map. If this step is Yes, the process proceeds to S1306. On the other hand, if this step is No, the process proceeds to S1310. In this embodiment, the image 1200 includes a personal information concealment area, so the process proceeds to S1306.

[0111] In S1306, the model output map conversion unit 322 performs image conversion on the output map in the image similarity calculation unit 302. The reason for performing image conversion on the output map is that personal information may remain in the output map. A detailed flow of S1306 is shown in Fig. 13(b).

[0112] In S1321, the image similarity calculation unit 302 calculates the image similarity between the original image and the output map. For example, the image similarity is calculated between the image 1200, which is the original image, and the pupil output map 1230, which is the output map. In this embodiment, the image similarity is calculated for each pupil region between the image 1200 and the pupil output map 1230. Here, the pupil region refers to the region in the labeled image 1210 to which the pupil labels 1212 and 1213 are assigned. Furthermore, the region used in the similarity calculation is not limited to a labeled region, and any region may be used, such as a hidden region or a region determined based on a hidden region. In this embodiment, it is assumed that the similarities of the regions to which the pupil labels 1212 and 1213 are assigned are 0.1 and 0.15, respectively.

[0113] In S1322, the model output map conversion parameter determination unit 323 acquires image conversion parameters corresponding to the similarity. In this embodiment, a filling process is performed when the image similarity is 0.5 or greater, and no image conversion is performed when the image similarity is less than 0.5. The method for determining the image conversion parameters is not limited to this; blurring or other image conversions may also be performed, and the strength of the image conversion may be changed depending on the similarity.

[0114] In S1323, the output map conversion unit 321 performs image conversion of the output map. Here, it is assumed that the similarities of the areas to which the pupil labels 1212 and 1213 are assigned are 0.1 and 0.15, respectively. In this case, since both similarities are less than 0.5, image conversion is not performed.

[0115] Thereafter, in S1324, the image similarity calculation unit 302 calculates the image similarity between the original image and the output map after the processing of S1323. Here, the image similarity between the image 1200 and the pupil output map after the processing of S1323 is calculated. Then, the image similarity calculation unit 302 determines whether the calculated image similarity is equal to or less than a threshold. If this step is Yes, the process ends. On the other hand, if this step is No, the process returns to S1322 and image conversion parameters are determined again. Here, the calculated image similarity is low and is equal to or less than the threshold, so the process ends and returns to the process of FIG. 13(a).

[0116] In S1307 of FIG. 13(a), the output map conversion unit 321 superimposes the image-converted output map on the personal information concealed image (for example, the personal information concealed image 1220 of FIG. 12).

[0117] In S1308, the output map conversion unit 321 superimposes a label on the image on which the output map has been superimposed. Here, it is assumed that the display image 1260 in FIG. 12 has been generated by superimposing pupil labels 1212 and 1213 on the image generated in S1307. The pupil output map 1230 and pupil labels 1212 and 1213 are superimposed on the personal information concealed image 1220. Note that, although an example in which label information corresponding to the output map being processed is superimposed has been described in the present embodiment, the present invention is not limited to this. Any label information may be assigned, and if there is no label information, no label information may be assigned.

[0118] In S1309, the output map conversion unit 321 determines whether all output maps have been processed. If the answer is Yes in this step, the process proceeds to S1311. On the other hand, if the answer is No in this step, the process returns to S1304 and continues.

[0119] In S1310, the learning unit 223 superimposes the output map on the original image. In S1311, the model output map display unit 331 updates the display screen using the generated display image, and displays the generated display image to user B. This completes the series of processes in FIG. 13(a).

[0120] Here, since there are output maps remaining that have not been processed in S1309, the process returns to S1304 and continues. Next, in S1304, the face output map 1240 is selected. Thereafter, the process proceeds to S1305 and S1306, where the process of FIG. 13(b) is performed. As with the pupil output map, it is assumed that the similarity between the image 1200 and the face output map 1240 is low. As a result, the display image 1270 of FIG. 12 is generated. Looking at the display image 1270, it can be seen that the face output map 1240 and the face label 1211 are superimposed on the personal information concealed image 1220. Thereafter, since there are output maps remaining that have not been processed, the process returns to S1304 and continues.

[0121] In S1304, the model intermediate output map 1250 is next selected. Then, the process proceeds to S1305 and S1306, where the processing of FIG. 13(b) is performed. In S1321, the image similarity of each concealed region between the image 1200 and the model intermediate output map 1250 is calculated. Here, the image similarity calculation unit 302 can calculate multiple similarities, such as edge similarity and color similarity. Any one of the multiple similarities, or the result with the highest numerical value, may be used as the image similarity. It is assumed that the similarity between the image 1200 and the model intermediate output map 1250 is low in terms of image color information but high in terms of image edge information, resulting in a calculated similarity of 0.8.

[0122] In S1322, image transformation parameters corresponding to the similarity are acquired. Here, since the image similarity (0.8) between the image 1200 and the model intermediate output map 1250 is equal to or greater than the threshold (0.5), it is determined that the hidden region should be filled in.

[0123] In S1323, a filling process is performed on the output map. Here, the area that has been filled is an area on the output map that corresponds to the area to which the concealed areas 1221 and 1222 of the personal information concealed image 1220 are assigned. The subsequent processing is the same as that described for the pupil output map, and therefore a description thereof will be omitted. Here, it is assumed that the display image 1280 of FIG. 12 has been generated. Looking at the display image 1280, it can be seen that the model intermediate output map 1250 is superimposed on the personal information concealed image 1220, and the concealed areas are filled in.

[0124] Thereafter, in S1309, since the processing of all the output maps has been completed, the process proceeds to S1311, and in S1311, the model output map display unit 331 displays the generated display image to the user B.

[0125] As described above, in this embodiment, even when checking the learning status of machine learning, it is possible to keep personal information confidential, and visual support can be provided to user B, who cannot see the original image.

[0126] In this embodiment, all learning tasks such as pupils and faces are described as tasks that can be displayed on the display screen, but this is not limited to this. For example, it may be determined whether or not to display the output map on the display screen depending on the label information assigned to the image. For example, it may be configured to display the output map for pupil labels and face labels, but not for mouth labels. Furthermore, displayable tasks may be set for each image.

[0127] [Variations] Although the above-described embodiments have been described using people as an example, other categories such as dogs, landscapes, etc. may also be targeted. Even when applying the above-described embodiments to other categories, learning can be performed using original images while keeping personal information confidential.

[0128] In the above embodiment, the case where there is only one information processing device shown in Fig. 2 has been described as an example, but the present invention is not limited to this. For example, user A and user B may each have an operation terminal equipped with a user interface unit 210 and an internal processing unit 220, and may perform operations using each operation terminal. In addition, the data management unit 230 may be a server device connected via a network, and the internal processing unit 220 may also be part of the information processing device managed on the network.

[0129] The disclosure of this specification includes the following information processing device, control method for an information processing device, and program.

[0130] (Item 1) a determining means for determining a concealment area for concealing personal information in the first image; a concealment unit that performs a concealment process on the first image based on the concealment area; a display means for displaying a second image, which is a concealed image that has been subjected to the concealment processing; a learning means for performing learning using the first image; An information processing device comprising:

[0131] (Item 2) further comprising an assigning means for assigning label information to the secret area; 2. The information processing device according to item 1, wherein the display means displays the second image and the label information.

[0132] (Item 3) 3. The information processing device according to item 2, wherein the display means displays the label information superimposed on the second image.

[0133] (Item 4) 4. The information processing device according to item 3, wherein the label information includes at least one region label from among a face label, an eye label, and a mouth label of a person included in the first image.

[0134] (Item 5) 3. The information processing device according to item 2, wherein the label information includes at least one classification label from among a category label, a gender label, and an orientation label of a person included in the first image.

[0135] (Item 6) 6. The information processing device according to any one of items 1 to 5, further comprising a management unit that manages a plurality of images including the first image and a plurality of confidential images including the second image.

[0136] (Item 7) further comprising an image similarity calculation means for calculating an image similarity between the secret area of ​​the first image and a secret area of ​​another image managed by the management means; 7. The information processing device according to item 6, wherein the display means acquires and displays images whose image similarity is equal to or greater than a threshold value from among the other images managed by the management means.

[0137] (Item 8) further comprising a label similarity calculation means for calculating a label similarity between label information assigned to the secret area of ​​the first image and label information assigned to other images managed by the management means; 8. The information processing device according to item 6 or 7, wherein the display means acquires and displays images whose label similarity is equal to or greater than a threshold value from among the other images managed by the management means.

[0138] (Item 9) further comprising a similar image generating means for converting the secret area into a similar image, 9. The information processing device according to any one of items 1 to 8, wherein the display means displays the second image in which the hidden area has been converted into the similar image.

[0139] (Item 10) a calculation unit for calculating an image similarity between the first image and the second image in which the hidden area is converted into the similar image; Item 9. The information processing device according to item 9, wherein the display means displays the second image in which the hidden area has been converted into the similar image when the image similarity is equal to or greater than a first threshold and equal to or less than a second threshold that is greater than the first threshold.

[0140] (Item 11) The first image is associated with learning utilization information; further comprising a score calculation means for calculating a recommendation score for learning use based on the learning use information of the first image; 11. The information processing device according to any one of items 1 to 10, wherein the display means further displays the learning use information and the learning use recommendation score.

[0141] (Item 12) Item 12. The information processing device according to item 11, wherein the learning utilization information includes at least one of the accuracy improvement rate when the first image is used for learning, the number of times the first image has been used for learning, and the number of likes for the first image.

[0142] (Item 13) 13. The information processing device according to any one of items 1 to 12, further comprising a map display means for displaying a model output map obtained as a result of inputting the first image into a model and performing learning.

[0143] (Item 14) further comprising a transformation means for transforming the model output map based on the hidden region of the first image; Item 14. The information processing apparatus according to item 13, wherein the map display means displays the model output map converted by the conversion means.

[0144] (Item 15) a calculation means for calculating a similarity between the first image and the model output map; a parameter determination means for determining a conversion parameter to be used by the conversion means so that the similarity is equal to or less than a threshold value; Item 15. The information processing device according to item 14, further comprising:

[0145] (Item 16) Item 16. The information processing device according to item 15, wherein the map display means determines whether or not to display the model output map according to label information assigned to the first image.

[0146] (Item 17) 17. The information processing device according to any one of items 1 to 16, wherein the display means displays the second image to the user when the user performs learning using the first image.

[0147] (Item 18) A control method for an information processing device, comprising: a determining step of determining a concealment area for concealing personal information in the first image; a concealment step of performing a concealment process on the first image based on the concealment area; a display step of displaying a second image, which is a concealed image obtained by the concealment process; a learning step of performing learning using the first image; 1. A method for controlling an information processing device, comprising:

[0148] (Item 19) A program for causing a computer to function as the information processing device according to any one of items 1 to 17.

[0149] (Other embodiments) The present invention can also be realized by supplying a program that realizes one or more functions of the above-described embodiments to a system or device via a network or a storage medium, and having one or more processors in the computer of the system or device read and execute the program.The present invention can also be realized by a circuit (e.g., ASIC) that realizes one or more functions.

[0150] The invention is not limited to the above-described embodiments, and various changes and modifications can be made without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0151] 200: Information processing device, 210: User interface unit, 211: Operation unit, 212: Display unit, 213: Data input unit, 220: Internal processing unit, 221: Concealed area determination unit, 222: Image conversion unit, 223: Learning unit, 224: Label assignment unit, 230: Data management unit, 231: Management unit, 232: Management data

Claims

1. a determining means for determining a concealment area for concealing personal information in the first image; a concealment unit that performs a concealment process on the first image based on the concealment area; a display means for displaying a second image, which is a concealed image that has been subjected to the concealment process; a learning means for performing learning using the first image; An information processing device comprising:

2. further comprising an assigning means for assigning label information to the secret area; 2. The information processing apparatus according to claim 1, wherein said display means displays said second image and said label information.

3. 3. The information processing apparatus according to claim 2, wherein the display means displays the label information superimposed on the second image.

4. The information processing apparatus according to claim 3 , wherein the label information includes at least one region label from among a face label, an eye label, and a mouth label of a person included in the first image.

5. The information processing apparatus according to claim 2 , wherein the label information includes at least one classification label selected from a category label, a gender label, and an orientation label of a person included in the first image.

6. 2. The information processing apparatus according to claim 1, further comprising a management unit that manages a plurality of images including the first image and a plurality of confidential images including the second image.

7. further comprising an image similarity calculation means for calculating an image similarity between the secret area of ​​the first image and a secret area of ​​another image managed by the management means; 7. The information processing apparatus according to claim 6, wherein the display means acquires and displays images whose image similarity is equal to or greater than a threshold value from among the other images managed by the management means.

8. a label similarity calculation unit that calculates a label similarity between label information assigned to the secret area of ​​the first image and label information assigned to other images managed by the management unit; 7. The information processing apparatus according to claim 6, wherein the display means acquires and displays images whose label similarity is equal to or greater than a threshold value from among the other images managed by the management means.

9. further comprising a similar image generating means for converting the secret area into a similar image, The information processing apparatus according to claim 1 , wherein the display unit displays the second image in which the hidden area is converted into the similar image.

10. a calculation unit for calculating an image similarity between the first image and the second image in which the hidden area is converted into the similar image; The information processing device according to claim 9, characterized in that the display means displays the second image in which the hidden area has been converted into the similar image when the image similarity is greater than or equal to a first threshold and less than or equal to a second threshold that is greater than the first threshold.

11. the first image is associated with learning utilization information; a score calculation means for calculating a recommendation score for learning use based on the learning use information of the first image; 2. The information processing apparatus according to claim 1, wherein the display means further displays the learning use information and the learning use recommendation score.

12. The information processing device according to claim 11, characterized in that the learning utilization information includes at least one of the accuracy improvement rate when the first image is used for learning, the number of times the first image has been used for learning, and the number of likes for the first image.

13. 2. The information processing apparatus according to claim 1, further comprising a map display means for displaying a model output map obtained as a result of inputting the first image into a model and performing learning.

14. a transforming means for transforming the model output map based on the hidden region of the first image; 14. The information processing apparatus according to claim 13, wherein said map display means displays the model output map converted by said conversion means.

15. a calculation means for calculating a similarity between the first image and the model output map; a parameter determination means for determining a conversion parameter to be used by the conversion means so that the similarity is equal to or less than a threshold value; 15. The information processing apparatus according to claim 14, further comprising:

16. 16. The information processing apparatus according to claim 15, wherein said map display means determines whether or not to display said model output map in accordance with label information assigned to said first image.

17. 2. The information processing apparatus according to claim 1, wherein the display means displays the second image to the user when the user performs learning using the first image.

18. A control method for an information processing device, comprising: a determining step of determining a concealment area for concealing personal information in the first image; a concealment step of performing a concealment process on the first image based on the concealment area; a display step of displaying a second image, which is a concealed image that has been subjected to the concealment process; a learning step of performing learning using the first image; 1. A method for controlling an information processing device, comprising:

19. A program for causing a computer to function as the information processing device according to any one of claims 1 to 17.

Images