Image determination device, image determination method, and image determination program

Abstract

To provide an image determination device, an image determination method, and an image determination program that enable an AI image to be detected with high accuracy.SOLUTION: An image determination device 1 includes: an image collection unit 11 that collects face images generated automatically by generative AI; an authentication processing unit 12 that performs predetermined face authentication processing for each of the collected face images and counts the number of other face images determined as faces identical to each of the face images; a master registration unit 13 that stores, as a masterpiece face, a face image counted by a predetermined number of times or more, among the collected face images; a determination unit 14 that determines whether an object image is a face identical to each of masterpiece faces through face authentication; and an output unit 15 that when there is a masterpiece face determined as an identical face, outputs a fact that the object image is an AI image as a determination result.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a technique for detecting images generated by AI.

Background Art

[0002] In recent years, a technique called generative AI, which automatically generates images and the like from specified descriptions, has been rapidly advancing. For this reason, for example, there has been pointed out the possibility of fraud such as impersonation and registration with non-existent personalities in online identity verification using face images generated by generative AI. Therefore, a system for discriminating between automatically generated images (AI images) and real photographic images has been developed (see, for example, Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in conventional systems for detecting AI images, methods such as using an AI that has learned AI images and non-AI images, or using the features of AI images (for example, blurring of the entire image) for determination are used, and false determinations are likely to occur. Therefore, further improvement in determination accuracy is desired.

[0005] An object of the present invention is to provide an image determination device, an image determination method, and an image determination program that can accurately detect AI images.

Means for Solving the Problems

[0006] The image determination device according to the present invention comprises: an image collection unit that collects face images automatically generated by a generation AI; an authentication processing unit that performs a predetermined face authentication process on each of the face images collected by the image collection unit and counts the number of other face images that are determined to be the same face as each face image; a master registration unit that stores as master faces the number of face images collected by the image collection unit that are counted by the authentication processing unit to a predetermined number or more; a determination unit that determines whether a target image is the same face as each of the master faces by face authentication; and an output unit that outputs as a determination result that the target image is an AI image if there is a master face that the determination unit determines to be the same face.

[0007] The output unit may output information indicating which of the Masupi faces was determined to be the same as the result of the determination.

[0008] The image collection unit may collect face images generated by multiple types of generation AI.

[0009] The image determination method according to the present invention includes an image collection step of collecting face images automatically generated by a generation AI; an authentication processing step of performing a predetermined face authentication process on each of the face images collected in the image collection step and counting the number of other face images determined to be the same face as each face image; a master registration step of storing as master faces the number of face images among the face images collected in the image collection step that are counted by the authentication processing step to a predetermined number or more; a determination step of determining whether a target image is the same face as each of the master faces by face authentication; and an output step of outputting as a determination result that the target image is an AI image if there are master faces determined to be the same face in the determination step.

[0010] The image determination program according to the present invention is for causing a computer to function as the image determination device. [Effects of the Invention]

[0011] According to the present invention, AI images can be detected with high accuracy. [Brief explanation of the drawing]

[0012] [Figure 1] This is a block diagram showing the functional configuration of the image determination device in the embodiment. [Figure 2] This flowchart shows the processing flow by the image determination device in the embodiment. [Modes for carrying out the invention]

[0013] An example of an embodiment of the present invention will be described below. The image recognition device of this embodiment stores a large number of face images extracted by face recognition from face images generated by various generation AIs as AI images, and determines whether the face image to be judged is an AI image or not by face recognition.

[0014] Figure 1 is a block diagram showing the functional configuration of the image determination device 1 in this embodiment. The image determination device 1 is an information processing device equipped with a control unit 10 and a storage unit 20, as well as various input / output interfaces.

[0015] The control unit 10 controls the entire image determination device 1 and realizes each function in this embodiment by appropriately reading and executing various programs stored in the storage unit 20. The control unit 10 may be a CPU. The storage unit 20 is a storage area for various programs and data that enable the hardware group to function as an image determination device 1, and may be ROM, RAM, flash memory, or a hard disk drive (HDD).

[0016] The control unit 10 functions as an image acquisition unit 11, an authentication processing unit 12, a master registration unit 13, a determination unit 14, and an output unit 15 by executing the software (image determination program) stored in the storage unit 20.

[0017] The image collection unit 11 collects face images automatically generated by a generative AI. These face images may be generated by the image determination device 1 itself or input from the outside. In addition, the image collection unit 11 may collect a variety of AI images by collecting face images generated by multiple types of generative AIs.

[0018] For each of the face images collected by the image collection unit 11, the authentication processing unit 12 performs a predetermined face authentication process of comparing it with other face images, and counts the number of other face images determined to be the same face as each face image.

[0019] The master registration unit 13 stores, in the storage unit 20, as masterpiece (masterpiece) faces, those face images among the face images collected by the image collection unit 11 for which the number counted by the authentication processing unit 12 is a predetermined value or more. Note that, among the face images determined to be the same face as each other, only one or a predetermined number (the top of the counted numbers) of them may be stored.

[0020] The determination unit 14 determines whether the input target image is the same face as each of the masterpiece faces by face authentication.

[0021] When there is a masterpiece face determined to be the same face by the determination unit 14, the output unit 15 outputs, as a determination result, that the target image is an AI image. In addition, the output unit 15 may also output, as a determination result, information on which masterpiece face it was determined to be the same face as.

[0022] FIG. 2 is a flowchart showing the processing flow by the image determination device 1 in the present embodiment. In step S1, the image collection unit 11 collects a large number (n) of face images generated by a generative AI or generates them itself, and designates them as M1 to Mn. In step S2, the authentication processing unit 12 performs a predetermined face recognition process for each of M1 to Mn and counts the number of images that are determined to be the same face as Mk (1 ≤ k ≤ n). In step S3, the master registration unit 13 stores as MP1 to MPm face images (m) where the number of identical faces counted in step S2 is greater than a certain amount.

[0023] In step S4, the determination unit 14 performs face recognition on each of the target images MP1 to MPm for which it is necessary to determine whether or not it is an AI image, and searches for the same face. In step S5, the output unit 15 determines whether or not there was a Masupi face that was recognized as identical in step S4. If the determination is YES, the process moves to step S6; if the determination is NO, the process moves to step S7. In step S6, the output unit 15 determines that the target image is an AI image automatically generated by the generation AI and outputs the determination result. In step S7, the output unit 15 determines that the target image is a non-AI image and outputs the determination result.

[0024] According to this embodiment, the image determination device 1 collects face images automatically generated by the generation AI and registers a mass-recognition face that has many other face images that are determined to be the same face by face recognition. As a result, when a target image is determined to be the same as a mass-recognition face by face recognition, the device can detect that the target image is an AI image. In other words, because it is difficult to input conditions for generating different people into the generating AI, the image recognition device 1 utilizes the property of generating a large number of similar face images called "mass faces," and by appropriately registering these mass faces, it can accurately detect that the target image is an AI image through face recognition. As a result, for example, it is possible to prevent with high accuracy the act of an attacker attempting to register a false identity using a facial image easily generated by AI.

[0025] Furthermore, the image recognition device 1 can acquire a Masupi face, which is a face that resembles an AI image, and when it determines that the target image is an AI image, it can clearly indicate which Masupi face it resembles and that a large number of images of that face have been generated. Therefore, the image recognition device 1 can show the user the reason why it determined the target image is an AI image, thereby building a highly reliable system.

[0026] Furthermore, the image recognition device 1 can comprehensively register Master Faces by collecting face images generated by multiple types of generation AI, thereby improving the detection accuracy of AI images.

[0027] Furthermore, this will help to curb fraudulent activities using AI images, for example, and thus contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs): "Build resilient infrastructure, promote sustainable industrialization and foster innovation."

[0028] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above. Furthermore, the effects described in the embodiments described above are merely a list of the most preferred effects resulting from the present invention, and the effects of the present invention are not limited to those described in the embodiments.

[0029] The image determination device 1 may utilize other determination techniques in addition to the image determination method of this embodiment. For example, images created using AI that trains on real people to generate similar faces are highly likely to be identified as those people by facial recognition technology. However, even such targeted AI images that recreate the faces of specific people can be detected by combining them with other technologies.

[0030] The image determination method by the image determination device 1 is implemented by software. When implemented by software, the programs constituting this software are installed on an information processing device (computer). These programs may be distributed to users by being recorded on removable media such as a CD-ROM, or by being downloaded to the user's computer via a network. Furthermore, these programs may be provided to the user's computer as a web service via a network without being downloaded. [Explanation of Symbols]

[0031] 1. Image determination device 10 Control Unit 11 Image Collection Department 12 Authentication Processing Unit 13 Master Registration Department 14 Judgment section 15 Output section 20 Memory section

Claims

1. An image collection unit that collects facial images automatically generated by a generation AI, An authentication processing unit performs a predetermined face recognition process on each of the face images collected by the image collection unit and counts the number of other face images that are determined to be the same face as each face image. A master registration unit stores, as master faces, the number of face images counted by the authentication processing unit that are equal to or greater than a predetermined number among the face images collected by the image collection unit, A determination unit that determines whether the target image is the same face as each of the aforementioned master faces by facial recognition, An image determination device comprising: an output unit that outputs as a determination result that the target image is an AI image when there is a mass face determined to be the same face by the determination unit.

2. The image determination device according to claim 1, wherein the output unit outputs information as a determination result indicating which of the Masupi faces the face was determined to be the same.

3. The image collection unit collects face images generated by multiple types of generation AI as described in claim 1 or claim 2.

4. An image collection step that collects face images automatically generated by a generation AI, A recognition processing step which involves performing a predetermined face recognition process on each of the face images collected in the image collection step, and counting the number of other face images that are determined to be the same face as each face image, A master registration step in which, among the face images collected in the image collection step, face images whose count in the authentication processing step is greater than or equal to a predetermined number are stored as master faces, A determination step of determining whether the target image is the same face as each of the aforementioned master faces by facial recognition, An image determination method in which a computer performs the following steps: an output step in which, if there is a Masupi face determined to be the same face in the determination step, the computer outputs as a determination result that the target image is an AI image.

5. An image determination program for causing a computer to function as an image determination device according to claim 1 or claim 2.

Images