Personal authentication apparatus, personal authentication method, and personal authentication program
The personal authentication device enhances accuracy by comparing face and non-face body part pulse waves, addressing the vulnerability of face authentication to forgery and ensuring genuine user intent.
Patent Information
- Application Number
- JP2024007793
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-01-23
- Publication Date
- 2025-08-04
AI Technical Summary
Existing face authentication technologies suffer from accuracy issues, particularly in the face of forgery attempts using photos or masks, necessitating improved personal authentication methods.
A personal authentication device that utilizes both an in-camera and an out-camera to capture face and non-face body parts, extracting pulse waves from these images and comparing their similarity to determine if they belong to the same individual, thereby enhancing authentication accuracy.
The method improves personal authentication accuracy by confirming the user's intent, ensuring that the authentication is performed willingly by the individual and not by someone else, even if face authentication is successfully forged.
Smart Images

Figure 2025113570000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a personal authentication device, a personal authentication method, and a personal authentication program.
Background Art
[0002] In recent years, face authentication, fingerprint authentication, or passwords (including PINs and patterns) have been used as personal authentication for identity verification.
[0003] On the other hand, face authentication has been advanced by AI (Artificial Intelligence) and can be used in a contactless, secure, and convenient manner. Therefore, face authentication is used for unlocking smartphones, managing access to rooms, etc.
[0004] However, such face authentication has vulnerabilities such as the lock function being unlocked by forgery using a face photo, video, artificial mask, etc. Therefore, as such a countermeasure, for example, a technique as described in Patent Document 1 has been proposed. The technique described in this Patent Document 1 detects facial movements such as eye blinks and mouth openings to determine whether it is a living body.
Prior Art Documents
Patent Documents
[0005]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0006] However, the above-described technique has a problem that there is still room for improvement in terms of accuracy.
[0007] Therefore, in view of the above problems, an object of the present invention is to provide a personal authentication device, a personal authentication method, and a personal authentication program that can improve the accuracy of personal authentication.
Means for Solving the Problem
[0008] The object of the present invention is achieved by the following means. Note that the reference numerals in parentheses are those of the embodiments described later, but the present invention is not limited thereto.
[0009] The personal authentication device according to claim 1 is a personal authentication device in which face authentication is performed, and includes a first pulse wave acquisition means (for example, the pulse wave extraction unit 14 shown in FIG. 3) that acquires the pulse wave of a face (for example, the face Hb of the person shown in FIG. 1), a second pulse wave acquisition means (for example, the pulse wave extraction unit 14 shown in FIG. 3) that acquires the pulse wave of a part other than the face, a confirmation means (for example, the pulse wave comparison determination unit 15 shown in FIG. 3) that confirms the similarity between the pulse wave acquired by the first pulse wave acquisition means and the pulse wave acquired by the second pulse wave acquisition means, and a determination means (for example, the pulse wave comparison determination unit 15 shown in FIG. 3) that determines whether or not the owners of the face and the part other than the face are the same person based on the similarity confirmed by the confirmation means.
[0010] The personal authentication device according to claim 2 is the personal authentication device according to claim 1, wherein the first pulse wave acquisition means (for example, the pulse wave extraction unit 14 shown in FIG. 3) acquires the pulse wave of a face (for example, the face Hb of the person shown in FIG. 1) by imaging using a built-in in-camera (for example, the in-camera 1b shown in FIG. 2), 2]the second pulse wave acquisition means (for example, the pulse wave extraction unit 14 shown in FIG. 3) acquires the pulse wave of the part other than the face from an image of the part other than the face imaged using a built-in out-camera (for example, the out-camera 1c shown in FIG. 2).
[0011] The personal authentication device according to claim 3 is the personal authentication device according to claim 1 or 2, wherein the determination means (for example, the pulse wave comparison determination unit 15 shown in FIG. 3) determines that the owner of the face and the part other than the face is the same person if the similarity confirmed by the confirmation means (for example, the pulse wave comparison determination unit 15 shown in FIG. 3) is within a predetermined threshold value set in advance.
[0012] The personal authentication method according to claim 4 is a personal authentication method using a predetermined device (for example, the personal authentication device 1 shown in FIG. 1) in which face authentication is performed, a step of acquiring a pulse wave of a face (for example, the face Hb of a person shown in FIG. 1) (for example, step S4 shown in FIG. 4), a step of acquiring a pulse wave of a part other than the face (for example, step S4 shown in FIG. 4), a step of confirming the similarity between the acquired pulse wave of the face and the acquired pulse wave of the part other than the face (for example, steps S5 to S7 shown in FIG. 4), and a step of determining whether or not the owner of the face and the part other than the face is the same person based on the confirmed similarity (for example, step S8 shown in FIG. 4).
[0013] The personal authentication program according to claim 5 is a personal authentication program in which face authentication is performed, causing a computer (for example, the personal authentication device 1 shown in FIG. 1) to execute a first process of acquiring a pulse wave of a face (for example, the face Hb of a person shown in FIG. 1) (for example, the pulse wave extraction unit 14 shown in FIG. 3), execute a second process of acquiring a pulse wave of a part other than the face (for example, the pulse wave extraction unit 14 shown in FIG. 3), execute a third process of confirming the similarity between the pulse wave acquired in the first process and the pulse wave acquired in the second process (for example, the pulse wave comparison determination unit 15 shown in FIG. 3), and execute a fourth process of determining whether or not the owner of the face and the part other than the face is the same person based on the similarity confirmed in the third process (for example, the pulse wave comparison determination unit 15 shown in FIG. 3).
Advantages of the Invention
[0014] Next, the effects of the present invention will be described with reference numerals in the drawings. The reference numerals in parentheses are those of the embodiments described later, but the present invention is not limited thereto.
[0015] According to the inventions according to claims 1, 4, and 5, by confirming the similarity of the pulse waves between the face and the parts other than the face, it is determined whether the owners of the face and the parts other than the face are the same person. Therefore, it is possible to determine whether face authentication is performed by one's own will rather than by others.
[0016] Therefore, according to the present invention, it is possible to improve the accuracy of personal authentication.
[0017] According to the invention according to claim 2, since the built-in in-camera (for example, the in-camera 1b shown in FIG. 2) and the built-in out-camera (for example, the out-camera 1c shown in FIG. 2) are used, there is no need to prepare a special sensor. Therefore, workability can be improved.
[0018] According to the invention according to claim 3, the accuracy of personal authentication can be further improved.
Brief Description of the Drawings
[0019]
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Modes for Carrying Out the Invention
[0020] Hereinafter, an embodiment of the personal authentication device according to the present invention will be specifically described with reference to the drawings. In the following description, when indicating the up, down, left, and right directions, it refers to the up, down, left, and right as viewed from the front shown in the drawing.
[0021] <Overview of the Personal Authentication Device> The personal authentication device according to this embodiment can improve the accuracy of personal authentication. Specifically, the personal authentication device 1 shown in FIG. 1 is composed of a smartphone or the like and can be carried by a person's hand Ha. Hereinafter, the personal authentication device 1 will be described in detail.
[0022] <Description of the Personal Authentication Device> As shown in FIG. 2, the personal authentication device 1 includes a CPU 1a that executes and controls each function of the personal authentication device 1, an in-camera 1b that is a camera disposed on the front surface side of a display 1g (see FIG. 1) included in the personal authentication device 1, an out-camera 1c that is a camera disposed on the back surface side of the display 1g (see FIG. 1) included in the personal authentication device 1, an input / output unit 1d that can input predetermined data into the personal authentication device 1 or output predetermined data outside the personal authentication device 1, a ROM 1e composed of a writable flash ROM or the like that stores a predetermined program or the like, a RAM 1f that functions as a work area, a buffer memory, etc., and a display 1g composed of an LCD (Liquid Crystal Display) or the like.
[0023] <Description of the Functional Configuration of the Personal Authentication Device> Next, the functional configuration of the personal authentication device 1 will be described with reference to FIG. 3. As shown in FIG. 3, the functional configuration of the personal authentication device 1 includes a determination unit 10, a storage unit 11, an image data acquisition unit 12, a face authentication unit 13, a pulse wave extraction unit 14, and a pulse wave comparison determination unit 15. Hereinafter, each configuration will be described.
[0024] The determination unit 10 exchanges information between each functional block based on the state held by the determination unit 10 or the state held by each functional block. Further, as shown in FIG. 3, the determination unit 10 also exchanges information between the in-camera 1b and the out-camera 1c.
[0025] The storage unit 11 stores data and the like necessary for processing.
[0026] The image data acquisition unit 12 acquires a face image of the person's face Hb shown in FIG. 1 and a finger image of the person's hand Ha shown in FIG. 1. Specifically, by turning the display 1g shown in FIG. 1 in the direction of the person's face, the in-camera 1b shown in FIG. 2 can capture the person's face Hb. Thereby, the image data acquisition unit 12 can acquire the face image of the person's face Hb shown in FIG. 1.
[0027] Furthermore, as shown in FIG. 1, by grasping the personal authentication device 1 with the person's hand Ha and turning the display 1g shown in FIG. 1 in the direction of the person's face Hb, the out-camera 1c shown in FIG. 2 can capture a finger image of the person's hand Ha. Thereby, the image data acquisition unit 12 can acquire the finger image of the person's hand Ha shown in FIG. 1.
[0028] The face authentication unit 13 authenticates whether the face image acquired by the image data acquisition unit 12 and the person pre-stored in the storage unit 11 are the same person. Since the face authentication unit 13 is a well-known technique in the art, detailed description thereof is omitted.
[0029] The pulse wave extraction unit 14 extracts the face pulse wave from the face image acquired by the image data acquisition unit 12, and further extracts the finger pulse wave from the finger image acquired by the image data acquisition unit 12.
[0030] The pulse wave comparison and determination unit 15 determines whether the face and the finger belong to the same person by comparing the face pulse wave extracted by the pulse wave extraction unit 14 and the finger pulse wave extracted by the pulse wave extraction unit 14.
[0031] <Description of an Example of Using a Personal Authentication Device> Next, with reference to the flowchart shown in FIG. 4, an example of using the personal authentication device 1 shown in this embodiment will be described. In the following, an example of a method for releasing the lock function of the personal authentication device 1 will be described.
[0032] First, when releasing the lock function of the personal authentication device 1, by turning the display 1g shown in FIG. 1 toward the direction of a person's face, the in-camera 1b shown in FIG. 2 captures the person's face Hb. In response to this, the determination unit 10 shown in FIG. 3 acquires the face image captured by the in-camera 1b and outputs it to the image data acquisition unit 12. As a result, the image data acquisition unit 12 acquires the face image captured by the in-camera 1b.
[0033] Furthermore, the out-camera 1c shown in FIG. 2 captures an image of the finger of the person's hand Ha. In response to this, the determination unit 10 shown in FIG. 3 acquires the finger image captured by the out-camera 1c and outputs it to the image data acquisition unit 12. As a result, the image data acquisition unit 12 acquires the finger image captured by the out-camera 1c (step S1).
[0034] Next, the determination unit 10 shown in FIG. 3 instructs the image data acquisition unit 12 to output the face image acquired by the image data acquisition unit 12 to the face authentication unit 13. In response to this, the image data acquisition unit 12 outputs the face image acquired to the face authentication unit 13. Furthermore, the determination unit 10 reads out the person image pre-stored in the storage unit 11 and outputs it to the face authentication unit 13. In response to this, the face authentication unit 13 compares the face image acquired by the image data acquisition unit 12 with the person image pre-stored in the storage unit 11 to authenticate whether they are the same person (step S2).
[0035] Next, the determination unit 10 shown in FIG. 3 receives the authentication result by the face authentication unit 13. If the face authentication is successful (step S3: YES), it proceeds to step S4. If the face authentication is not successful (step S3: NO), the lock function of the personal authentication device 1 is not released, and a series of processes are terminated.
[0036] Next, when face authentication is successful (step S3: YES), the determination unit 10 shown in FIG. 3 instructs the image data acquisition unit 12 to output the face image and finger image acquired by the image data acquisition unit 12 to the pulse wave extraction unit 14. In response to this, the image data acquisition unit 12 outputs the face image and finger image acquired by it to the pulse wave extraction unit 14. Thereby, the pulse wave extraction unit 14 analyzes the image data of the face image and finger image acquired by the image data acquisition unit 12 for each frame (step S4). More specifically, focusing on the characteristic that hemoglobin absorbs green light, the pulse wave extraction unit 14 analyzes the face image and finger image acquired by the image data acquisition unit 12 for each frame for the light change that varies due to the contraction and expansion of blood vessels. Specifically, in the face image, the pulse wave extraction unit 14 acquires the G (green) value of each RGB value of the left cheek part, and in the finger image, the central part of the finger, and applies a band-pass filter (for example, 0.7 to 1.7 Hz) to extract the time-series data of the face pulse wave and finger pulse wave. Thereby, pulse wave data as shown in FIG. 5 can be extracted.
[0037] Next, the determination unit 10 shown in FIG. 3 instructs the pulse wave extraction unit 14 to output the pulse wave data extracted by the pulse wave extraction unit 14 to the pulse wave comparison and determination unit 15. In response to this, the pulse wave extraction unit 14 outputs the pulse wave data extracted by it to the pulse wave comparison and determination unit 15. Thereby, the pulse wave comparison and determination unit 15 detects peaks and valleys from the pulse wave data extracted by the pulse wave extraction unit 14 (step S5). Specifically, in order not to detect false peaks and valleys due to noise, the amplitude magnitude H is set in advance as the minimum detection value for peaks and valleys, and based on this, the pulse wave comparison and determination unit 15 detects peaks and valleys from the pulse wave data extracted by the pulse wave extraction unit 14 whose amplitude magnitude (value) is H or more. The result is as shown in FIG. 5. Specifically, the "〇" mark in FIG. 5 indicates a peak, and the "▽" mark indicates a valley.
[0038] Next, the pulse wave comparison determination unit 15 forms a pair in which the detected time difference P between peaks and the detected time difference V between troughs is minimized (step S6). Specifically, for each peak and trough of the facial pulse wave data, the pulse wave comparison determination unit 15 calculates P and V for all peaks and troughs of the finger pulse wave data, and forms a pair when P and V are minimized and are less than or equal to the maximum allowable value M of the time difference. The result is the portion indicated by the dashed line in FIG. 5.
[0039] Next, the pulse wave comparison determination unit 15 calculates the average value Pave of P and the average value Vave of V within a certain period of the frame shown in FIG. 5, and sets the sum value to T. That is, T = Pave + Vave is calculated. Then, the pulse wave comparison determination unit 15 instructs the determination unit 10 to read out the threshold value D stored in advance in the storage unit 11. In response to this, the determination unit 10 reads out the threshold value D stored in advance in the storage unit 11 and outputs it to the pulse wave comparison determination unit 15. Thereby, the pulse wave comparison determination unit 15 determines whether the sum value T is less than or equal to the threshold value D (step S7). As a specific value of this threshold value D, for example, when the in-camera 1b captures a person's face Hb at 24 fps for 5 seconds, and at the same time, the out-camera 1c captures an image of a person's finger Ha at 24 fps for 5 seconds, 5.6 frames (= 5.6 / 24 ≈ 0.23 seconds) is exemplified as the value of the threshold value D.
[0040] Thus, if the sum value T is less than or equal to the threshold value D (step S8: YES), the pulse wave comparison determination unit 15 determines that the owners of the face and the finger are the same person, and outputs the result to the determination unit 10. In response to this, the determination unit 10 performs a process of releasing the lock function of the personal authentication device 1 (step S9) and ends a series of processes.
[0041] On the other hand, if the sum value T is not less than or equal to the threshold value D (step S8: NO), the pulse wave comparison determination unit 15 determines that the owners of the face and the finger are not the same person, and outputs the result to the determination unit 10. In response to this, the determination unit 10 does not release the lock function of the personal authentication device 1 and ends a series of processes.
[0042] Thus, even if face authentication is successfully achieved through forgery, if the face and the owner of the finger are not the same person, the lock function of the personal authentication device 1 cannot be released. That is, according to this embodiment, it is possible to determine whether the user is trying to release the lock by their own will rather than by someone else.
[0043] Therefore, according to this embodiment, the accuracy of personal authentication can be improved.
[0044] <Description of Modification Example> Note that the shapes and the like shown in this embodiment are merely examples, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims. For example, in this embodiment, an example is shown in which the face pulse wave is extracted by imaging the human face Hb with the in-camera 1b, and the finger pulse wave is extracted by imaging the finger image of the human hand Ha with the out-camera 1c. However, it is not necessary to use the in-camera 1b and the out-camera 1c. For example, when imaging the human face Hb or the finger image of the human hand Ha, an external camera not built into the personal authentication device 1 may be used for imaging. Also, when extracting the pulse wave, the pulse wave may be extracted using a pulse wave sensor or the like without using a camera. That is, any method may be used as long as the pulse wave can be extracted. However, it is more preferable to extract the face pulse wave by imaging the human face Hb with the in-camera 1b and extract the finger pulse wave by imaging the finger image of the human hand Ha with the out-camera 1c because the in-camera 1b and the out-camera 1c built into a smartphone or the like can be used, and there is no need to prepare a special sensor. Therefore, it is preferable because the workability can be improved.
[0045] In addition, in this embodiment, an example was shown in which if the total value T is equal to or less than the threshold value D, it is determined that the face and the owner of the finger are the same person. However, the present invention is not limited to this, and a lower limit value of the threshold value may be provided, and if it is within a predetermined threshold value range, it may be determined that the face and the owner of the finger are the same person. By doing so, even if some problem occurs and the total value T is near 0, it will not be determined that the face and the owner of the finger are the same person, so that the accuracy of personal authentication can be further improved.
[0046] In addition, in this embodiment, an example was shown in which when the appearance time difference is equal to or less than the maximum allowable value M, a pair in which the appearance time difference P between the detected peaks and the appearance time difference V between the detected valleys becomes the minimum is formed. However, the present invention is not limited to this, and a lower limit value of the maximum allowable value M of the appearance time difference may be provided. By doing so, since it is unlikely that the time difference between the peaks and valleys of the pulse waves at two different measurement sites (exemplified as the face and the finger in this embodiment) will be 0 (if the measurement sites are different, the distances from the heart are different, so an appearance time difference will naturally occur), if a lower limit value of the maximum allowable value M or less is provided, the accuracy of personal authentication can be improved.
[0047] In addition, in this embodiment, when comparing the face pulse wave data and the finger pulse wave data, the method of steps S5 to S8 shown in FIG. 4 was exemplified. However, the present invention is not limited to this, and the period of the pulse wave data may be collated, and the similarity of the feature portions may be performed using methods such as the Manhattan distance and the Euclidean distance. That is, the method for the similarity of the pulse wave data is not particularly limited.
[0048] In addition, in this embodiment, an example of extracting the finger pulse wave was shown. However, any pulse wave at a location other than the face, such as the arm or leg, may be used.
[0049] In addition, in this embodiment, as the personal authentication device 1, a portable smartphone was exemplified. However, the present invention is not limited to this, and it may be applied to a fixed device that cannot be carried.
[0050] In addition, in this embodiment, an example of releasing the lock function by face authentication is shown. However, the present invention is not limited thereto, and it is also applicable to entry and exit management of buildings and offices where face authentication is introduced, identity verification at banks, and ticket sales.
[0051] Moreover, as the personal authentication device 1 exemplified in this embodiment, it may be provided as the personal authentication device 1 as in this embodiment, or may be installed on the cloud and accessed from a user terminal. Further, it may be provided as an application program and installed on an existing smartphone, PC, etc. so that it can be used as the personal authentication device 1.
Explanation of Reference Numerals
[0052] 1 Personal authentication device (predetermined device, computer) 1b In-camera 1c Out-camera 14 Pulse wave extraction unit (first pulse wave acquisition means, second pulse wave acquisition means, first process, second process) 15 Pulse wave comparison and determination unit (confirmation means, determination means, third process, fourth process)
Claims
1. In a personal authentication device that performs face authentication, a first pulse wave acquisition means for acquiring a face pulse wave; a second pulse wave acquisition means for acquiring a pulse wave of a part other than the face; a confirmation means for confirming the similarity between the pulse wave acquired by the first pulse wave acquisition means and the pulse wave acquired by the second pulse wave acquisition means; a determination means for determining whether the owners of the face and the part other than the face are the same person based on the similarity confirmed by the confirmation means. A personal authentication device comprising:
2. The first pulse wave acquisition means acquires a face pulse wave from an image of a face captured using a built-in in-camera, The second pulse wave acquisition means acquires a pulse wave of a part other than the face from an image of a part other than the face captured using a built-in out-camera. The personal authentication device according to claim 1.
3. If the similarity confirmed by the confirmation means is within a predetermined threshold value set in advance, the determination means determines that the owners of the face and the part other than the face are the same person. The personal authentication device according to claim 1 or 2.
4. A personal authentication method using a predetermined device for performing face authentication, comprising: a step of acquiring a face pulse wave; a step of acquiring a pulse wave of a part other than the face; a step of confirming the similarity between the acquired face pulse wave and the acquired pulse wave of a part other than the face; a step of determining whether the owners of the face and the part other than the face are the same person based on the confirmed similarity. A personal authentication method comprising:
5. In a personal authentication program for performing face authentication, causing a computer to perform a first process of acquiring a face pulse wave; perform a second process of acquiring a pulse wave of a part other than the face; perform a third process of confirming the similarity between the pulse wave acquired by the first process and the pulse wave acquired by the second process; A personal authentication program for causing the computer to perform a fourth process of determining whether the owners of the face and the part other than the face are the same person based on the similarity confirmed by the third process.
Citation Information
Patent Citations
Face authentication device
JP2006330936A