Face anomaly detection method and device, equipment and storage medium

Abstract

The invention discloses a face anomaly detection method and device, equipment and a storage medium, and belongs to the technical field of face recognition. The method comprises the steps of firstly obtaining to-be-detected face data; performing living body detection based on the to-be-detected face data to obtain a living body detection result; and if the living body detection result is qualified, performing face recognition detection based on the to-be-detected face data, and executing corresponding processing operation according to the obtained face recognition detection result. According to the invention, the living body detection and the face recognition detection are successively carried out based on the obtained to-be-detected face data, so that the face recognition detection result is finally obtained and the corresponding processing operation is executed, and effective detection of the abnormal face condition is realized.

Description

Technical Field

[0001] The present invention relates to the field of face recognition technology, and in particular to a face anomaly detection method, device, equipment and storage medium. Background Art

[0002] With the rapid development of science and technology and the advantages of face recognition technology itself, such as non-invasiveness and security, face recognition technology has been more and more widely used, such as mobile phone unlocking, face payment, etc. However, in real life, online fraud and illegal acts of inducing or using other people's bank cards to withdraw money are becoming more and more complicated and difficult to prevent, especially when ATM machines cannot be provided with 24-hour manual monitoring and prevention, an offline anti-fraud system is urgently needed to detect and warn operators of facial anomalies.

[0003] In view of this, the present invention proposes a complete system for facial anomaly detection, which can perform anomaly detection on facial data, and accurately identify the facial state and issue a warning in time by judging the anomaly of the position of facial key points.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a method, device, equipment and storage medium for detecting facial anomalies, aiming to solve the technical problem that the prior art cannot effectively detect facial anomalies.

[0006] To achieve the above object, the present invention provides a method for detecting facial anomalies, the method comprising the following steps:

[0007] Get the face data to be detected;

[0008] Performing liveness detection based on the face data to be detected to obtain a liveness detection result;

[0009] If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result.

[0010] Optionally, before obtaining the face data to be detected, the following steps are further included:

[0011] Get parameter setting instructions;

[0012] Adjusting the model parameters of the trained face anomaly detection model based on the parameter setting instruction to obtain an adjusted face anomaly detection model;

[0013] The adjusted face anomaly detection model is deployed in an edge computing device to obtain a face anomaly detection device.

[0014] Optionally, performing liveness detection based on the face information to be detected to obtain a liveness detection result specifically includes:

[0015] Analyzing the face data to be detected to obtain depth information of the face to be detected;

[0016] Liveness detection is performed based on the depth information of the face to be detected to obtain the liveness detection result.

[0017] Optionally, if the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result, specifically including:

[0018] If the liveness detection result is qualified, performing face recognition detection based on the face data to be detected to obtain a face recognition detection result;

[0019] If the face recognition test result is qualified, face information is extracted based on the face data to be detected, and the obtained face recognition information is stored and displayed;

[0020] If the face recognition detection result is unqualified, a face detection abnormality feedback report is output based on the face recognition detection result.

[0021] Optionally, if the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected to obtain a face recognition detection result, which specifically includes:

[0022] If the liveness detection result is qualified, performing facial key point existence detection based on the face data to be detected to obtain a first face data detection result;

[0023] If the first face data detection result is qualified, facial key point position detection is performed based on the face data to be detected to obtain the face recognition detection result.

[0024] Optionally, if the first face data detection result is qualified, facial key point position detection is performed based on the face data to be detected to obtain the face recognition detection result, specifically including:

[0025] If the first face data detection result is qualified, then performing facial key point relative position detection based on the face data to be detected to obtain a second face data detection result;

[0026] If the detection result of the second face data is qualified, relative distance detection of facial key points is performed based on the face data to be detected to obtain the face recognition detection result.

[0027] Optionally, after performing liveness detection based on the face data to be detected to obtain a liveness detection result, the method further includes:

[0028] If the liveness detection result is unqualified, a liveness detection abnormality report is generated based on the liveness detection result, and the face data to be detected is reacquired for subsequent data detection.

[0029] In addition, to achieve the above-mentioned purpose, the present invention also proposes a face anomaly detection device, the face anomaly detection device comprising:

[0030] Data acquisition module: obtains the face data to be detected;

[0031] Liveness detection module: performs liveness detection based on the face data to be detected to obtain a liveness detection result;

[0032] Identification and detection module: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result.

[0033] In addition, to achieve the above-mentioned purpose, the present invention also proposes a face anomaly detection device, which includes: a memory, a processor, and a face anomaly detection program stored in the memory and runnable on the processor, and the face anomaly detection program is configured to implement the steps of the face anomaly detection method described above.

[0034] In addition, to achieve the above-mentioned purpose, the present invention also proposes a computer-readable storage medium storing a computer program, wherein a face anomaly detection program is stored on the storage medium, and when the face anomaly detection program is executed by a processor, the steps of the face anomaly detection method described above are implemented.

[0035] The present invention first obtains face data to be detected; performs liveness detection based on the face data to be detected to obtain liveness detection results; if the liveness detection results are qualified, performs face recognition detection based on the face data to be detected, and performs corresponding processing operations according to the obtained face recognition detection results. The present invention performs liveness detection and face recognition detection in sequence based on the acquired face data to be detected, thereby finally obtaining face recognition detection results and performing corresponding processing operations, thereby achieving effective detection of facial abnormalities. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a structural diagram of a face anomaly detection device in a hardware operating environment involved in an embodiment of the present invention;

[0037] Figure 2It is a flowchart of the first embodiment of the method for detecting abnormal face in the present invention;

[0038] Figure 3 It is a flowchart of the second embodiment of the method for detecting abnormal face in the present invention;

[0039] Figure 4 It is a flowchart of a third embodiment of the method for detecting facial anomalies of the present invention;

[0040] Figure 5 It is a structural block diagram of the first embodiment of the face anomaly detection device of the present invention.

[0041] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0042] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0043] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of a face anomaly detection device in the hardware operating environment involved in the embodiment of the present invention.

[0044] like Figure 1 As shown, the face anomaly detection device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0045] Those skilled in the art will understand that Figure 1 The structure shown in does not constitute a limitation on the facial anomaly detection device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0046] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a face anomaly detection program.

[0047] exist Figure 1 In the face anomaly detection device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the face anomaly detection device of the present invention can be set in the face anomaly detection device, and the face anomaly detection device calls the face anomaly detection program stored in the memory 1005 through the processor 1001, and executes the face anomaly detection method provided by the embodiment of the present invention.

[0048] The embodiment of the present invention provides a method for detecting facial anomalies. Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of a method for detecting facial anomalies according to the present invention.

[0049] In this embodiment, the face anomaly detection method comprises the following steps:

[0050] Step S10: Acquire face data to be detected;

[0051] It should be noted that, in a specific implementation, the face data to be detected is obtained by scanning the user using an imaging device. For example, a three-dimensional depth image of the user's face (face data to be detected) can be obtained through a TOF camera (Time-of-Flight Camera).

[0052] Step S20: performing liveness detection based on the face data to be detected to obtain a liveness detection result;

[0053] It should be noted that, in a specific implementation, the purpose of the liveness detection operation is to exclude two-dimensional face data such as photos and videos through depth information, thereby ensuring that the identified object is alive.

[0054] Step S30: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result.

[0055] It is understandable that in the specific implementation, when it is confirmed that the identification object is a living body (the user really exists), it is necessary to further perform face recognition detection on the face data to be detected, wherein the purpose of face recognition detection is to effectively obtain the face information of the identification object (user) and ensure the integrity of the information. For example, ensure that the user's face is not blocked (such as wearing a mask to block the mouth and nose, wearing sunglasses to block the eyes and pupils) or ensure that abnormal situations such as artificially tilting the face to avoid information collection are excluded.

[0056] This embodiment first obtains the face data to be detected; performs liveness detection based on the face data to be detected to obtain a liveness detection result; if the liveness detection result is qualified, performs face recognition detection based on the face data to be detected, and performs corresponding processing operations according to the obtained face recognition detection result. This embodiment performs liveness detection and face recognition detection based on the acquired face data to be detected, thereby finally obtaining a face recognition detection result and performing corresponding processing operations, thereby achieving effective detection of facial abnormalities.

[0057] refer to Figure 3 , Figure 3 The figure is a flow chart of a second embodiment of a method for detecting facial anomalies according to the present invention.

[0058] Based on the first embodiment above, in this embodiment, before step S10, the following steps are further included:

[0059] Step S00: Obtain parameter setting instructions;

[0060] It can be understood that, in a specific implementation, the parameter setting instruction is issued by the system administrator based on the user interface of the face anomaly detection system, which can be used to adjust the parameters of the face anomaly detection system.

[0061] Step S01: adjusting model parameters of the trained face anomaly detection model based on the parameter setting instruction to obtain an adjusted face anomaly detection model;

[0062] It should be noted that in the specific implementation, the system administrator can adjust and set the various parameters of the face anomaly detection system through parameter setting instructions based on usage requirements. For example, the user can set algorithm-related parameters on the settings page, such as the face anomaly detection confidence threshold (generally set to 0.7) to control the credibility of the final output of normal face detection; scene category parameter selection such as long-distance (within 3 meters) and close-distance (within 0.5 meters) face information capture, etc., to issue different scene model weights.

[0063] Step S02: deploy the adjusted face anomaly detection model to the edge computing device to obtain a face anomaly detection device.

[0064] It should also be noted that in the specific implementation, the face anomaly detection model includes a face anomaly detection algorithm, which can be used to perform face recognition detection on the face data to be detected. The edge computing device specifically refers to an edge box (edge ​​computing platform) with GPU computing capabilities, which can process video streams (face data to be detected) in real time and trigger anomaly detection, thereby improving the response speed and real-time performance of the face anomaly detection system.

[0065] After obtaining the parameter setting instruction, this embodiment can adjust the model parameters of the trained face anomaly detection model according to the parameter setting instruction, and then deploy the adjusted face anomaly detection model to the edge computing device to obtain the face anomaly detection equipment, which provides a basis for subsequent face anomaly detection.

[0066] Furthermore, liveness detection is performed based on the face information to be detected to obtain a liveness detection result, specifically including: parsing the face data to be detected to obtain depth information of the face to be detected; and performing liveness detection based on the depth information of the face to be detected to obtain a liveness detection result.

[0067] It should be noted that, in the specific implementation, facial depth information specifically refers to the extraction and analysis of the three-dimensional structural information of the face through deep learning technology, including the positioning of facial feature points, changes in facial expressions, and the geometric shape of the face.

[0068] refer to Figure 4 , Figure 4 The figure is a flow chart of a third embodiment of a method for detecting facial anomalies according to the present invention.

[0069] Based on the above first embodiment, in this embodiment, step S30 specifically includes:

[0070] Step S31: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected to obtain a face recognition detection result;

[0071] It should be noted that, in the specific implementation, only when the liveness detection result is qualified (the detected object / user is alive), face recognition detection will be performed based on the face data to be detected.

[0072] It should also be noted that, in a specific implementation, if the liveness detection result is unqualified, the recognition detection process of the face data to be detected will be terminated, and the user's face will be captured again to obtain new face data to be detected.

[0073] Step S32: If the face recognition detection result is qualified, face information is extracted based on the face data to be detected, and the obtained face recognition information is stored and displayed;

[0074] It can be understood that in the specific implementation, the face recognition detection operation is implemented based on a three-step filtering logic based on the position judgment of facial key points, which can accurately identify the abnormal state of the face, and may specifically include facial key point existence detection, facial key point position detection and facial key point relative position detection.

[0075] It should be noted that, in the specific implementation, facial key points refer to the facial feature positions preset in the face recognition detection model, such as eyebrows, eyes, nose, mouth and facial contours. In this embodiment, the specific objects of facial key points can be adjusted and set by the model manager according to usage requirements.

[0076] Step S33: If the face recognition detection result is unqualified, a face detection abnormality feedback report is output based on the face recognition detection result.

[0077] It is understandable that, in a specific implementation, an unqualified face recognition detection result specifically refers to a situation where the detected user has face occlusion or face tilt (facial recognition angle tilt) and other situations that make it impossible to perform normal recognition detection.

[0078] It should also be noted that in the specific implementation, the face detection abnormality feedback report can be displayed and queried through the device user interface, so as to understand the specific reasons for the detection abnormality (such as facial occlusion, tilted facial recognition angle, and too long facial recognition distance, etc.).

[0079] In this embodiment, when the liveness detection result is qualified, face recognition detection can be first performed based on the face data to be detected, and then corresponding data processing operations can be performed according to the face recognition detection result to complete the face detection operation.

[0080] Furthermore, if the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected to obtain a face recognition detection result, which specifically includes: if the liveness detection result is qualified, facial key point existence detection is performed based on the face data to be detected to obtain a first face data detection result; if the first face data detection result is qualified, facial key point position detection is performed based on the face data to be detected to obtain a face recognition detection result.

[0081] It should be noted that, in a specific implementation, the specific process of performing facial key point presence detection based on the face data to be detected is to first read the preset facial key point data (for example, left / right eyebrows, left / right eyes and mouth), and then perform facial key point presence detection on the face detection frame. When the face data to be detected meets the standard requirements of all preset facial key points (the user's left / right eyebrows, left / right eyes and mouth can be detected from the face data to be detected), it can be determined that the detection is qualified and the first face data detection result (qualified detection) is obtained; conversely, when the face data to be detected does not meet the standard requirements of all preset facial key points, it can be determined that the detection is unqualified and the first face data detection result (unqualified detection) is obtained.

[0082] It should also be noted that in the specific implementation, in this embodiment, the face anomaly detection model can specifically extract face features through the CNN convolutional neural network, use the Wing Loss function and change its loss calculation method to perform additional loss calculation on abnormal key points, so that the occluded key points can be regressed to the position of the original face, while not affecting the normal regression of the unoccluded key points. The Wing Loss function is defined as:

[0083]

[0084] Among them, |yy^| is the absolute error between the predicted value and the true value (regression error); δ is a smoothing threshold that controls whether to use the smoothing term when the error is small (can be set to 1.0); α controls the growth rate of the loss (can be set to 1.2), and β controls the weight of the loss when the error is large (can be set to 2.5).

[0085] It can be understood that in this embodiment, the CNN convolutional network model used is specifically RetinaFace (a facial detection algorithm based on deep learning). At the same time, for the own edge device of this embodiment, this embodiment adopts a lightweight optimization operation to achieve normal reasoning on the box (the specific operation is to use Mobilenet (a lightweight convolutional neural network) with extremely small model parameters to embed the meta-model for implementation); and because the model's accuracy will be lost after the model is lightweight, the Squeeze-and-Excitation (a convolutional neural network structure) channel attention mechanism is added to the lightweight model structure to ensure that the extent of the decrease in model accuracy does not affect the effect of the model.

[0086] Furthermore, if the first face data detection result is qualified, facial key point position detection is performed based on the face data to be detected to obtain the face recognition detection result, specifically including: if the first face data detection result is qualified, facial key point relative position detection is performed based on the face data to be detected to obtain the second face data detection result; if the second face data detection result is qualified, facial key point relative distance detection is performed based on the face data to be detected to obtain the face recognition detection result.

[0087] It should be noted that in the specific implementation, the role of the relative position detection of facial key points based on the face data to be detected is mainly to check whether the relative position of each facial key point conforms to the geometric structure of a normal face, for example, the key points on both sides of the two eyes and the mouth should be on a horizontal line, and the nose should be located between the eyes and the mouth; if the relative position of the five key points does not meet the standard, an abnormal warning is returned. The specific process can be to first read the preset relative position data of facial key points (for example, the relative position between the left / right eyebrows, the left / right eyes and the mouth), and then perform relative position detection of facial key points on the face detection frame.

[0088] It should also be noted that in the specific implementation, the role of the relative distance detection of facial key points based on the face data to be detected is mainly to accurately determine whether the distance between the mouth, eyes and nose is reasonable by calculating the relative distance between the five key points, and to ensure that the distance difference does not exceed the reasonable range (for example, the difference in the y-axis distance between the left and right corners of the mouth and the nose should not exceed 50% of the maximum distance). If the conditions are not met, it is determined to be abnormal and trigger an early warning. The specific process can be to first read the preset relative distance data of facial key points (for example, the relative distance between the left / right eyebrows can be set to 2-5 cm), and then perform relative distance detection of facial key points on the face detection frame.

[0089] It is understandable that in a specific implementation, if the first face data detection result is unqualified or the second face data detection result is unqualified, the detection of the face data to be detected will be terminated and the abnormal detection result will be sent to the user interface for storage and display.

[0090] Furthermore, after performing liveness detection based on the face data to be detected to obtain a liveness detection result, it also includes: if the liveness detection result is unqualified, generating a liveness detection abnormality report based on the liveness detection result, and re-acquiring the face data to be detected for subsequent data detection.

[0091] In addition, an embodiment of the present invention also proposes a computer-readable storage medium storing a computer program, wherein a face anomaly detection program is stored on the storage medium, and when the face anomaly detection program is executed by a processor, the steps of the face anomaly detection method described above are implemented.

[0092] Since the storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described one by one here.

[0093] Reference Figure 5 , Figure 5 It is a structural block diagram of the first embodiment of the face anomaly detection device of the present invention.

[0094] like Figure 5 As shown, the face anomaly detection device proposed in the embodiment of the present invention includes:

[0095] Data acquisition module 10: acquiring face data to be detected;

[0096] Liveness detection module 20: performs liveness detection based on the face data to be detected to obtain liveness detection results;

[0097] Identification and detection module 30: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result.

[0098] This embodiment first obtains the face data to be detected; performs liveness detection based on the face data to be detected to obtain a liveness detection result; if the liveness detection result is qualified, performs face recognition detection based on the face data to be detected, and performs corresponding processing operations according to the obtained face recognition detection result. This embodiment performs liveness detection and face recognition detection based on the acquired face data to be detected, thereby finally obtaining a face recognition detection result and performing corresponding processing operations, thereby achieving effective detection of facial abnormalities.

[0099] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0100] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0101] In addition, for technical details not described in detail in this embodiment, reference can be made to the face anomaly detection method provided in any embodiment of the present invention, and will not be repeated here.

[0102] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0103] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0105] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for detecting facial anomalies, characterized in that: include: Get the face data to be detected; Performing liveness detection based on the face data to be detected to obtain a liveness detection result; If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result.

2. The method for detecting facial anomalies according to claim 1, wherein: Before obtaining the face data to be detected, the following steps are also included: Get parameter setting instructions; Adjusting the model parameters of the trained face anomaly detection model based on the parameter setting instruction to obtain an adjusted face anomaly detection model; The adjusted face anomaly detection model is deployed in an edge computing device to obtain a face anomaly detection device.

3. The method for detecting facial anomalies according to claim 1, wherein: Performing liveness detection based on the face information to be detected to obtain a liveness detection result specifically includes: Analyzing the face data to be detected to obtain depth information of the face to be detected; Liveness detection is performed based on the depth information of the face to be detected to obtain the liveness detection result.

4. The method for detecting facial anomalies according to any one of claims 1 to 3, characterized in that: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result, specifically including: If the liveness detection result is qualified, performing face recognition detection based on the face data to be detected to obtain a face recognition detection result; If the face recognition test result is qualified, face information is extracted based on the face data to be detected, and the obtained face recognition information is stored and displayed; If the face recognition detection result is unqualified, a face detection abnormality feedback report is output based on the face recognition detection result.

5. The method for detecting facial anomalies according to claim 4, characterized in that: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected to obtain a face recognition detection result, which specifically includes: If the liveness detection result is qualified, performing facial key point existence detection based on the face data to be detected to obtain a first face data detection result; If the first face data detection result is qualified, facial key point position detection is performed based on the face data to be detected to obtain the face recognition detection result.

6. The method for detecting facial anomalies according to claim 5, characterized in that: If the first face data detection result is qualified, then the facial key point position detection is performed based on the face data to be detected to obtain the face recognition detection result, which specifically includes: If the first face data detection result is qualified, then performing facial key point relative position detection based on the face data to be detected to obtain a second face data detection result; If the detection result of the second face data is qualified, relative distance detection of facial key points is performed based on the face data to be detected to obtain the face recognition detection result.

7. The method for detecting facial anomalies according to any one of claims 1 to 3, characterized in that: After performing liveness detection based on the face data to be detected to obtain a liveness detection result, the method further includes: If the liveness detection result is unqualified, a liveness detection abnormality report is generated based on the liveness detection result, and the face data to be detected is reacquired for subsequent data detection.

8. A facial anomaly detection device, characterized in that: The face anomaly detection device comprises: Data acquisition module: obtains the face data to be detected; Liveness detection module: performs liveness detection based on the face data to be detected to obtain a liveness detection result; Identification and detection module: If the liveness detection result is qualified, face recognition detection is performed based on the face data to be detected, and corresponding processing operations are performed according to the obtained face recognition detection result.

9. A facial anomaly detection device, characterized in that: The face anomaly detection device comprises: a memory, a processor, and a face anomaly detection program stored in the memory and executable on the processor, wherein the face anomaly detection program is configured to implement the face anomaly detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps in the facial anomaly detection method described in any one of claims 1 to 7 can be implemented.

Images