Face authenticity recognition detection method, system, device and storage medium

Abstract

The present application relates to the technical field of image recognition, and particularly relates to a face authenticity recognition detection method, system, device and storage medium, the method comprising the following steps: S1, when collecting a face image, recording a pattern position and a click operation position and corresponding first time node information, and collecting face image information; S2, analyzing human eye sight line change information according to the face image information, and then comparing and verifying the analysis, if the position and time are both verified to be consistent, then generating a first authenticity recognition verification passed prompt; S3, calling a camera to collect gesture information, and second time node information corresponding to the gesture information; S4, comparing and verifying the gesture information, and verifying whether the two time node information is consecutive, if the verification results are both passed, then generating a second authenticity recognition verification passed prompt; S5, determining a face authenticity recognition result according to the results of the two authenticity recognition verification passed prompts. The present application can improve the precision of face image authenticity recognition.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, specifically to methods, systems, devices, and storage media for detecting and verifying the authenticity of faces. Background Technology

[0002] With the development of facial recognition technology, facial recognition systems have been widely used in applications such as access control systems, login systems, mobile payments, bank transfers, and real-name dating apps. However, in the facial recognition image acquisition stage, some people attach paper strips with their images to their foreheads to circumvent facial recognition authentication. Others use image editing tools to replace their ID card photos with other portrait images, then use the software's path function to outline features such as eyebrows, eyes, nose, and mouth. They then use scripts to convert static images into dynamic videos of blinking, head turning, and mouth opening, and store these videos in a corresponding folder on a mobile emulator. Bypassing the camera, the video is directly transmitted to the detection program, ultimately fooling liveness detection and successfully bypassing facial recognition authentication, thus leading to low security in facial recognition systems. Summary of the Invention

[0003] One of the objectives of this invention is to provide a method for detecting the authenticity of faces, which can improve the accuracy of face image authenticity detection, thereby enhancing the security of the face recognition system.

[0004] To achieve the above objectives, a method for detecting the authenticity of faces is provided, including the following steps:

[0005] S1. When collecting facial images, a random pattern is set on the collection interface of the collection terminal, and a corresponding tap prompt is set. After tapping the pattern, the position of the pattern changes randomly. The pattern position, the tap operation position, and the corresponding first time node information are recorded. At the same time, the facial image information of the operator is collected through the front camera of the collection terminal.

[0006] S2. Analyze the changes in human eye gaze based on the facial image information, and then compare and verify the changes in human eye gaze with the recorded pattern position, tap operation position and first time node information. If the position and time are verified to be consistent, a first authenticity verification prompt is generated.

[0007] S3. After capturing the face image, set the gesture pose prompt on the capture interface of the capture terminal, and call the rear or front camera of the capture terminal to capture the gesture information of the operator according to the gesture pose prompt, as well as the second time node information of the corresponding gesture information.

[0008] S4. Compare and verify the gesture information with the gesture pose prompt, and verify whether the information at the first time node and the information at the second time node are continuous. If the verification result is that the gestures are the same and the information at the first time node and the information at the second time node are continuous, then generate a second authenticity verification prompt.

[0009] S5. Based on the prompts indicating that the first and second authenticity verifications have passed, determine the result of the face authenticity verification.

[0010] Furthermore, step S1 also includes the following steps:

[0011] S101. When acquiring a face image, extract the face image frame from it, analyze the brightness information of the face based on the face image frame, and control the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information. Record the exposure adjustment time node information of the exposure brightness adjustment, and finally adjust the brightness to the specified brightness threshold.

[0012] Step S2 also includes the following steps:

[0013] S201. Before analyzing the changes in human eye gaze based on the face image information, find the corresponding image frame from the face image information based on the exposure adjustment time node information, analyze whether the image frame has the set exposure change, if so, mark the face image information as reliable, and continue the process of step S2.

[0014] Furthermore, in step S3, the gestures used for posing include rock, paper, scissors, thumb and index finger pinching, thumb and middle finger pinching, and thumb and ring finger pinching; step S3 further includes the following steps:

[0015] S301. Randomly select several gestures from the gestures, and extract corresponding gesture posing prompt example images from the database;

[0016] S302. Set the execution order of several gestures, and randomly set the display time of the example image for each gesture pose prompt.

[0017] The second objective of this invention is to provide a face authenticity recognition and detection system, comprising the following modules:

[0018] First data acquisition module: When acquiring face images, it sets a random pattern on the acquisition interface of the acquisition terminal and sets a corresponding tap prompt. After tapping the pattern, the position of the pattern changes randomly. It records the pattern position, the tap operation position, and the corresponding first time node information. At the same time, it acquires the operator's face image information through the front-facing camera of the acquisition terminal.

[0019] The first comparison and verification analysis module is used to analyze the changes in human eye gaze based on the facial image information, and then compare and verify the changes in human eye gaze with the recorded pattern position, tap operation position and first time node information. If the position and time are verified to be consistent, a prompt indicating that the first authenticity recognition verification has passed is generated.

[0020] The second data acquisition module is used to set gesture prompts on the acquisition interface of the acquisition terminal after acquiring face images, and to call the rear or front camera of the acquisition terminal to acquire the gesture information of the operator according to the gesture prompts, as well as the second time node information of the corresponding gesture information.

[0021] The second comparison and verification analysis module is used to compare and verify the gesture information with the gesture pose prompt, and to verify whether the first time node information and the second time node information are continuous. If the verification result is that the gestures are the same and the first time node information and the second time node information are continuous, then a second authenticity verification pass prompt is generated.

[0022] The authenticity verification result analysis module is used to determine the authenticity of a face based on the first and second authenticity verification pass prompts.

[0023] Furthermore, the first data acquisition module also includes the following sub-modules:

[0024] Exposure brightness adjustment submodule: When capturing face images, it extracts face image frames, analyzes the brightness information of the face based on the face image frames, and controls the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information. It records the exposure adjustment time node information and finally adjusts the brightness to the specified brightness threshold.

[0025] The first comparison and verification analysis module also includes the following sub-modules:

[0026] The primary facial credibility analysis submodule is used to find the corresponding image frame from the facial image information based on the exposure adjustment time node information before analyzing the change information of human eye gaze based on the facial image information. It analyzes whether the image frame has the set exposure change. If it does, the facial image information is marked as credible and the process of the first comparison and verification analysis module continues.

[0027] Furthermore, in the second data acquisition module, the gesture prompts include rock, paper, scissors, thumb and index finger pinching, thumb and middle finger pinching, and thumb and ring finger pinching; the second data acquisition module includes the following sub-modules:

[0028] The gesture random extraction submodule is used to randomly extract several gestures from the gesture library and extract corresponding gesture pose prompt example images from the database.

[0029] Gesture Settings Submodule: Used to set the execution order of several gestures, and to randomly set the display time of the example image for each gesture pose prompt.

[0030] A third objective of this invention is to provide an electronic device, including a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the face authenticity detection method described above are performed.

[0031] The fourth objective of this invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the face authenticity detection method described above.

[0032] Principles and advantages:

[0033] 1. When acquiring facial images, first extract facial image frames, analyze the brightness information of the face based on the facial image frames, and control the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information, recording the exposure adjustment time node information; then, before analyzing the change information of the human eye gaze based on the facial image information, find the corresponding image frame from the facial image information based on the exposure adjustment time node information, and analyze whether the image frame has the set exposure change. If it does, the facial image information is marked as reliable. If it is a directly imported video stream, there is no exposure brightness. Therefore, random 1-2 exposure brightness adjustments are used to prove that the camera of the acquisition terminal is acquiring facial images normally.

[0034] 2. Furthermore, in this solution, when capturing facial images, the acquisition terminal's interface displays randomly positioned patterns. Users need to tap these patterns, and the system records the pattern's position, the tap's location, and the corresponding first time point information. Simultaneously, the front-facing camera of the acquisition terminal captures the operator's facial image information. Then, based on the facial image information, the system analyzes changes in the user's eye gaze, comparing these changes with the recorded pattern positions, tap locations, and first time point information. If both the position and time are verified to be consistent, a first verification success message is generated. This demonstrates that the acquisition terminal's camera is normally capturing real human facial images, preventing the use of paper strips with human images pasted on foreheads to circumvent facial authenticity verification.

[0035] 3. Furthermore, after acquiring a facial image, this solution sets a gesture prompt on the acquisition terminal's interface and uses the terminal's rear or front camera to capture the operator's gestures according to the prompt, along with the corresponding second time point information. The gesture information is then compared and analyzed with the gesture prompt, and the continuity between the first and second time point information is verified. If the verification result shows identical gestures and continuity between the first and second time point information, a second verification success message is generated. The continuity between the first and second time point information is crucial; discontinuity indicates the presence of remote video sharing and recombination techniques to deceive the system, while continuity confirms the authenticity of the face. Through these three stages of authenticity verification, the accuracy of facial image authenticity recognition is improved, thereby enhancing the security of the facial recognition system. Attached Figure Description

[0036] Figure 1 This is a flowchart of the face authenticity detection method according to an embodiment of the present invention. Detailed Implementation

[0037] The following detailed description illustrates the specific implementation method:

[0038] Example

[0039] A method for detecting the authenticity of a face, basically as follows: Figure 1 As shown, it includes the following steps:

[0040] S1. When acquiring facial images, a randomly positioned pattern is set on the acquisition interface of the acquisition terminal, and a corresponding tap prompt is set. After tapping the pattern, the position of the pattern changes randomly. The pattern position, the tap operation position, and the corresponding first time node information are recorded. At the same time, the facial image information of the operator is acquired through the front-facing camera of the acquisition terminal. Step S1 also includes the following steps:

[0041] S101. When acquiring face images, extract face image frames, analyze the brightness information of the face based on the face image frames, and control the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information. Record the exposure adjustment time node information, and finally adjust the brightness to the specified brightness threshold; for example, first lower the exposure brightness to darken the entire image, then increase the exposure brightness to brighten the entire image, and finally adjust it to a normal brightness that can detect the face. This is to prove that the camera of the acquisition terminal is acquiring face images normally. If it is a pre-set video, there will be no corresponding brightness changes at the time nodes.

[0042] Before performing step S2, the following steps are also included:

[0043] S201. Before analyzing the changes in human eye gaze based on the face image information, find the corresponding image frame from the face image information based on the exposure adjustment time node information, analyze whether the image frame has the set exposure change, if so, mark the face image information as reliable, and continue the process of step S2.

[0044] S2. Analyze the changes in human eye gaze based on the facial image information, and then compare and verify the changes in human eye gaze with the recorded pattern position, tap operation position, and first time node information. If the position and time are verified to be consistent, a first authenticity verification prompt is generated. The pattern position represents the pixel display area on the mobile phone screen, and the tap operation position represents the finger tap position on the mobile phone screen. Both have a meaningful correspondence with world time. During the comparison and verification analysis, both the position and time must be compared.

[0045] When the first verification prompt is generated, it indicates that the camera of the data acquisition terminal is normally capturing images of real people's faces, preventing paper strips with images of people from being pasted on the forehead to avoid circumventing the verification of face authenticity.

[0046] S3. After capturing the facial image, a gesture prompt is set on the capture interface of the capture terminal, and the rear or front camera of the capture terminal is used to capture the gesture information of the operator according to the gesture prompt, as well as the second time node information of the corresponding gesture information; in step S3, the gesture prompt includes rock, paper, scissors, thumb and index finger pinch, thumb and middle finger pinch, and thumb and ring finger pinch; step S3 further includes the following steps:

[0047] S301. Randomly select several gestures from the gestures and extract corresponding gesture posing prompt example images from the database; for example, randomly select two gestures from the gestures, namely scissors and thumb and index finger pinching, and extract the corresponding gesture posing prompt example images from the database.

[0048] S302. Set the execution order of several gestures, and randomly set the display time of the example image for each gesture. The execution order of the gestures is: scissors, thumb and forefinger pinching, and each gesture is displayed for 2 seconds.

[0049] S4. Compare and verify the gesture information with the gesture pose prompt, and verify whether the information at the first time node and the information at the second time node are continuous. If the verification result is that the gestures are the same and the information at the first time node and the information at the second time node are continuous, then generate a second authenticity verification pass prompt. The continuity between the information at the first time node and the information at the second time node is the key point. If they are not continuous, it means that there is a technical means of remote video sharing and re-synthesis (there is a time difference) to deceive the system. If they are continuous, it means that it is a real person.

[0050] S5. Based on the successful verification prompts for both the first and second verifications (both with successful prompts), determine the result of face authentication. This solution can eliminate methods of face authentication fraud, such as pasting a piece of paper on the forehead or synthesizing face videos. Furthermore, the above three-stage authentication process improves the accuracy of face image authentication, thereby enhancing the security of the face recognition system.

[0051] A facial recognition and detection system includes a server and a data acquisition terminal, which are remotely connected. The data acquisition terminal is a conventional smartphone. The server includes the following modules:

[0052] The first data acquisition module is used to set a randomly positioned pattern on the acquisition interface of the acquisition terminal when acquiring facial images, and to set a corresponding tap prompt. After tapping the pattern, the position of the pattern changes randomly. The module records the pattern position, the tap position, and the corresponding first time point information. Simultaneously, it acquires the operator's facial image information through the front-facing camera of the acquisition terminal. The first data acquisition module also includes the following sub-modules:

[0053] Exposure brightness adjustment submodule: When capturing face images, it extracts face image frames, analyzes the brightness information of the face based on the face image frames, and controls the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information. It records the exposure adjustment time node information and finally adjusts the brightness to the specified brightness threshold.

[0054] The first comparison and verification analysis module is used to analyze changes in human eye gaze based on facial image information, and then compare and verify these changes with recorded pattern positions, tap operation positions, and first time node information. If the positions and times are consistent, a first authenticity verification success message is generated. The first comparison and verification analysis module also includes the following sub-modules:

[0055] The primary facial credibility analysis submodule is used to find the corresponding image frame from the facial image information based on the exposure adjustment time node information before analyzing the change information of human eye gaze based on the facial image information. It analyzes whether the image frame has the set exposure change. If it does, the facial image information is marked as credible and the process of the first comparison and verification analysis module continues.

[0056] The second data acquisition module is used to set gesture prompts on the acquisition interface of the acquisition terminal after acquiring facial images, and to call the rear or front camera of the acquisition terminal to capture the gesture information of the operator according to the gesture prompts, as well as the second time node information of the corresponding gesture information; in the second data acquisition module, the gesture prompts include rock, paper, scissors, thumb and index finger pinch, thumb and middle finger pinch, and thumb and ring finger pinch; the second data acquisition module includes the following sub-modules:

[0057] The gesture random extraction submodule is used to randomly extract several gestures from the gesture library and extract corresponding gesture pose prompt example images from the database.

[0058] Gesture Settings Submodule: Used to set the execution order of several gestures, and to randomly set the display time of the example image for each gesture pose prompt.

[0059] The second comparison and verification analysis module is used to compare and verify the gesture information with the gesture pose prompt, and to verify whether the first time node information and the second time node information are continuous. If the verification result is that the gestures are the same and the first time node information and the second time node information are continuous, then a second authenticity verification pass prompt is generated.

[0060] The authenticity verification result analysis module is used to determine the authenticity of a face based on the first and second authenticity verification pass prompts.

[0061] An electronic device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the face authenticity detection method described above are performed.

[0062] A computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the face authenticity detection method described above.

[0063] Those skilled in the art will understand that implementing all or part of the above-described face authenticity detection method can be accomplished by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the various embodiments of the face authenticity detection method described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0064] The above descriptions are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, based on the guidance provided in this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for detecting the authenticity of a face, characterized in that, Includes the following steps: S1. When collecting facial images, a random pattern is set on the collection interface of the collection terminal, and a corresponding tap prompt is set. After tapping the pattern, the position of the pattern changes randomly. The pattern position, the tap operation position, and the corresponding first time node information are recorded. At the same time, the facial image information of the operator is collected through the front camera of the collection terminal. S2. Analyze the changes in human eye gaze based on the facial image information, and then compare and verify the changes in human eye gaze with the recorded pattern position, tap operation position and first time node information. If the position and time are verified to be consistent, a first authenticity verification prompt is generated. S3. After capturing the face image, set the gesture pose prompt on the capture interface of the capture terminal, and call the rear or front camera of the capture terminal to capture the gesture information of the operator according to the gesture pose prompt, as well as the second time node information of the corresponding gesture information. S4. Compare and verify the gesture information with the gesture pose prompt, and verify whether the information at the first time node and the information at the second time node are continuous. If the verification result is that the gestures are the same and the information at the first time node and the information at the second time node are continuous, then generate a second authenticity verification prompt. S5. Based on the prompts indicating that the first and second authenticity verifications have passed, determine the result of the face authenticity verification.

2. The face authenticity detection method according to claim 1, characterized in that: Step S1 also includes the following steps: S101. When acquiring a face image, extract the face image frame from it, analyze the brightness information of the face based on the face image frame, and control the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information. Record the exposure adjustment time node information of the exposure brightness adjustment, and finally adjust the brightness to the specified brightness threshold. Step S2 also includes the following steps: S201. Before analyzing the changes in human eye gaze based on the face image information, find the corresponding image frame from the face image information based on the exposure adjustment time node information, analyze whether the image frame has the set exposure change, if so, mark the face image information as reliable, and continue the process of step S2.

3. The face authenticity detection method according to claim 2, characterized in that: In step S3, the gesture prompts include rock, paper, scissors, thumb and index finger pinching, thumb and middle finger pinching, and thumb and ring finger pinching; step S3 further includes the following steps: S301. Randomly select several gestures from the gestures, and extract corresponding gesture posing prompt example images from the database; S302. Set the execution order of several gestures, and randomly set the display time of the example image for each gesture pose prompt.

4. A face authenticity recognition and detection system, characterized in that, Includes the following modules: First data acquisition module: When acquiring face images, it sets a random pattern on the acquisition interface of the acquisition terminal and sets a corresponding tap prompt. After tapping the pattern, the position of the pattern changes randomly. It records the pattern position, the tap operation position, and the corresponding first time node information. At the same time, it acquires the operator's face image information through the front-facing camera of the acquisition terminal. The first comparison and verification analysis module is used to analyze the changes in human eye gaze based on the facial image information, and then compare and verify the changes in human eye gaze with the recorded pattern position, tap operation position and first time node information. If the position and time are verified to be consistent, a prompt indicating that the first authenticity recognition verification has passed is generated. The second data acquisition module is used to set gesture prompts on the acquisition interface of the acquisition terminal after acquiring face images, and to call the rear or front camera of the acquisition terminal to acquire the gesture information of the operator according to the gesture prompts, as well as the second time node information of the corresponding gesture information. The second comparison and verification analysis module is used to compare and verify the gesture information with the gesture pose prompt, and to verify whether the first time node information and the second time node information are continuous. If the verification result is that the gestures are the same and the first time node information and the second time node information are continuous, then a second authenticity verification pass prompt is generated. The authenticity verification result analysis module is used to determine the authenticity of a face based on the first and second authenticity verification pass prompts.

5. The face authenticity recognition and detection system according to claim 4, characterized in that: The first data acquisition module also includes the following sub-modules: Exposure brightness adjustment submodule: When capturing face images, it extracts face image frames, analyzes the brightness information of the face based on the face image frames, and controls the camera of the acquisition terminal to randomly adjust the exposure brightness 1-2 times based on the brightness information. It records the exposure adjustment time node information and finally adjusts the brightness to the specified brightness threshold. The first comparison and verification analysis module also includes the following sub-modules: The primary facial credibility analysis submodule is used to find the corresponding image frame from the facial image information based on the exposure adjustment time node information before analyzing the change information of human eye gaze based on the facial image information. It analyzes whether the image frame has the set exposure change. If it does, the facial image information is marked as credible and the process of the first comparison and verification analysis module continues.

6. The face authenticity recognition and detection system according to claim 5, characterized in that: In the second data acquisition module, the gesture prompts include rock, paper, scissors, thumb and index finger pinching, thumb and middle finger pinching, and thumb and ring finger pinching; the second data acquisition module includes the following sub-modules: The gesture random extraction submodule is used to randomly extract several gestures from the gesture library and extract corresponding gesture pose prompt example images from the database. Gesture Settings Submodule: Used to set the execution order of several gestures, and to randomly set the display time of the example image for each gesture pose prompt.

7. An electronic device, characterized in that: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions that the processor can execute. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of the face authenticity detection method as described in any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the face authenticity detection method as described in any one of claims 1 to 3.

Images