Liveness detection method, system and storage medium

Abstract

The present invention provides a liveness detection method, system and storage medium, which belong to the field of artificial intelligence technology, and specifically include: determining the depth information between different feature points and different reference points of a user's facial image through a binocular camera device, and determining the depth information credibility of different feature points of the user and the deviation feature points in combination with preset depth information, determining the initial liveness recognition probability of the user through the depth information credibility of different feature points and the deviation feature points, when the initial liveness recognition probability does not meet the requirements, determining the number of identity authentication failures of the user within a preset time based on the user's identity information, and determining the liveness recognition probability and liveness recognition result of the user through the depth information credibility of different feature points with different identity authentication failure times and the deviation feature points, thereby improving the accuracy of liveness recognition.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and in particular relates to a living body detection method, system and storage medium. Background Art

[0002] In order to realize liveness detection in the process of face recognition, the existing technical solution CN202011217244.7 "Liveness detection method, device and storage medium based on binocular depth and image structure information" deducts images of different face proportions, and respectively calculates the mean difference of depth information between M key points corresponding to face features in the image and N reference points to realize the recognition of liveness. However, at the same time, there are the following technical problems:

[0003] The simple mean difference of depth information cannot accurately realize the recognition of non-living objects. In some special occasions, 3D masks made of various materials can also be used to make certain depth information exist between the key points and reference points of different images. Therefore, the recognition of non-living objects by the above means may result in the recognition accuracy of living objects failing to meet the requirements.

[0004] In order to solve the above technical problems, the present invention provides a living body detection method, system and storage medium. Summary of the invention

[0005] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0006] According to one aspect of the present invention, a living body detection method is provided.

[0007] A method for detecting a living body, characterized by comprising:

[0008] S1 determines the identity information of the user, and determines the preset depth information between different feature points of the user and different reference points according to the identity information;

[0009] S2 determines the depth information between different feature points of the user's facial image and different reference points through a binocular camera device, and determines the depth information credibility of different feature points of the user and the deviation feature points in combination with the preset depth information, determines the initial liveness recognition probability of the user through the depth information credibility of the different feature points and the deviation feature points, and proceeds to the next step when the initial liveness recognition probability does not meet the requirement;

[0010] S3: acquiring the infrared image of the user's face, and determining that the infrared living body recognition probability of the user meets the requirement based on the deviation between the temperature characteristics of different characteristic areas of the infrared image of the face and the historical temperature characteristics of different characteristic areas of the user, and then proceeding to the next step;

[0011] S4 determines the number of identity authentication failures of the user within a preset time based on the identity information of the user, and determines the user's liveness recognition probability and liveness recognition result through the depth information credibility of different feature points with different identity authentication failure times and deviation feature points.

[0012] The beneficial effects of the present invention are:

[0013] 1. The user's initial liveness recognition probability is determined by the credibility of the depth information of different feature points and the deviation feature points, thereby avoiding the technical problem of inaccurate liveness recognition results caused by simply using the deviation of the depth information. By comparing with the preset depth information, it is further combined with the user's identity information to improve the accuracy of liveness recognition.

[0014] 2. The user's infrared liveness recognition probability is determined based on the deviation between the temperature characteristics of different characteristic areas of the facial infrared image and the historical temperature characteristics of different characteristic areas of the user, and then further compared with the user's identity information from the perspective of temperature characteristics, thereby achieving accurate recognition of disguised users such as 3D face masks and ensuring the accuracy of liveness recognition results.

[0015] 3. The user's liveness recognition probability and liveness recognition result are determined by the depth information credibility of different feature points with different numbers of identity authentication failures and the deviation feature points, thereby avoiding the influence of the angle deviation of the user's facial image on the accuracy of the liveness recognition result in a single identity authentication, and further improving the accuracy of the liveness recognition result.

[0016] A further technical solution is that the identity information of the user is determined based on a recognition result of the user's facial image, specifically based on a matching result between the user's facial image and a facial image database.

[0017] A further technical solution is that the preset depth information is determined based on the depth information of the reserved facial image of the user in the facial image database or based on the depth information of the facial image of the user during historical identity authentication.

[0018] A further technical solution is that when the credibility of the depth information of the feature point does not meet the requirement, the feature point is determined to be a deviation feature point.

[0019] A further technical solution is to determine whether the credibility of the depth information of the user's feature points meets the requirements, specifically including:

[0020] When the credibility of the depth information of the feature points of the user is greater than a preset credibility threshold, it is determined that the credibility of the depth information of the feature points of the user meets the requirement.

[0021] A further technical solution is that the liveness recognition result of the user includes liveness and non-liveness.

[0022] In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that: the processor executes the above-mentioned liveness detection method when running the computer program.

[0023] In a third aspect, the present invention provides a computer storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the above-mentioned method for detecting a living body.

[0024] Other features and advantages will be described in the following description. The objects and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and drawings.

[0025] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0027] Figure 1 is a flow chart of a liveness detection method;

[0028] Figure 2 is a flow chart of a method for determining the credibility of depth information of a feature point;

[0029] Figure 3 It is a framework diagram of a computer storage medium. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0031] Method Example

[0032] To solve the above problems, according to one aspect of the present invention, Figure 1 As shown, a living body detection method is provided, which is characterized by specifically comprising:

[0033] S1 determines the identity information of the user, and determines the preset depth information between different feature points of the user and different reference points according to the identity information;

[0034] It should be noted that the identity information of the user is determined based on the recognition result of the user's facial image, and specifically based on the matching result between the user's facial image and a facial image database.

[0035] Furthermore, the preset depth information is determined according to depth information of a reserved facial image of the user in the facial image database or according to depth information of a facial image of the user during historical identity authentication.

[0036] S2 determines the depth information between different feature points of the user's facial image and different reference points through a binocular camera device, and determines the depth information credibility of different feature points of the user and the deviation feature points in combination with the preset depth information, determines the initial liveness recognition probability of the user through the depth information credibility of the different feature points and the deviation feature points, and proceeds to the next step when the initial liveness recognition probability does not meet the requirement;

[0037] In one possible embodiment, Figure 2 As shown, the method for determining the credibility of the depth information of the feature point in the above step S2 is:

[0038] Determine the deviation of the depth information between the feature point and different reference points through the depth information between the feature point and different reference points and the difference of the preset depth information, and judge whether there is a reference point whose deviation does not meet the requirement, if yes, proceed to the next step, if not, determine the credibility of the depth information of the feature point through the average value of the deviation of the depth information between the feature point and different reference points;

[0039] The reference point whose deviation meets the requirement is taken as a credible reference point, and whether the reference point is credible is determined according to the number of the credible reference points and their proportion in the reference points. If so, the credibility of the depth information of the feature point is determined by the average value of the deviation of the depth information between the feature point and different reference points. If not, proceed to the next step;

[0040] The reference point whose deviation does not meet the requirement is taken as the deviation reference point, and the comprehensive deviation of the deviation reference point is determined according to the number of the deviation reference points, the maximum value and the average value of the deviation of the deviation reference point, and it is judged whether the comprehensive deviation of the deviation reference point meets the requirement, if so, proceed to the next step, if not, determine the credibility of the depth information of the feature point through the comprehensive deviation;

[0041] The credibility of the depth information of the feature point is determined based on the number of the credible reference points, the average value of the deviation of the depth information between the feature point and different reference points, and the comprehensive deviation.

[0042] Specifically, when the credibility of the depth information of the feature point does not meet the requirement, the feature point is determined to be a deviation feature point.

[0043] In a possible embodiment, the method for determining the initial liveness recognition probability of the user in the above step S2 is:

[0044] S21 determines the maximum value of the depth information credibility of the user's feature points based on the depth information credibility of the different feature points, and judges whether the maximum value of the depth information credibility of the user's feature points meets the requirement, if so, proceeds to the next step, if not, determines that the initial liveness recognition probability of the user does not meet the requirement;

[0045] S22 obtains the number of deviation feature points of the user, and determines whether the initial liveness recognition probability of the user does not meet the requirement based on the proportion of the number of deviation feature points to the number of feature points of the user. If so, it is determined that the initial liveness recognition probability of the user does not meet the requirement. If not, proceed to the next step.

[0046] S23 determines the credibility of the depth information of the deviation feature points of the user based on the deviation feature points of the user, and determines the comprehensive deviation amount of the deviation feature points of the user in combination with the number of the deviation feature points of the user and the proportion of the number of the deviation feature points in the number of the feature points of the user, and determines whether the initial liveness recognition probability of the user does not meet the requirement based on the comprehensive deviation amount of the deviation feature points of the user, and if so, determines that the initial liveness recognition probability of the user does not meet the requirement, and if not, proceeds to the next step;

[0047] S24 takes the feature points of the user whose depth information credibility is greater than the preset credibility as credible feature points, and determines the initial liveness recognition probability of the user according to the number of the user's credible feature points, the depth information credibility of different credible feature points, the comprehensive deviation of the user's deviation feature points, and the number of feature points.

[0048] It should be noted that judging whether the credibility of the depth information of the user's feature points meets the requirements specifically includes:

[0049] When the credibility of the depth information of the feature points of the user is greater than a preset credibility threshold, it is determined that the credibility of the depth information of the feature points of the user meets the requirement.

[0050] In another possible embodiment, the method for determining the initial liveness recognition probability of the user in step S2 is:

[0051] When the credibility of the depth information of the feature points of the user meets the requirements, it is determined that the initial liveness recognition probability of the user meets the requirements, and the initial liveness recognition probability is determined by the average value of the credibility of the depth information of the feature points of the user;

[0052] When the user has feature points whose depth information credibility does not meet the requirements, determine whether the initial liveness recognition probability of the user does not meet the requirements based on the number of deviation feature points of the user and the proportion of the number of deviation feature points to the number of feature points of the user. If so, determine that the initial liveness recognition probability of the user does not meet the requirements. If not, proceed to the next step.

[0053] Determine the credibility of the depth information of the deviation feature points of the user based on the deviation feature points of the user, and determine the comprehensive deviation amount of the deviation feature points of the user in combination with the number of the deviation feature points of the user and the proportion of the number of the deviation feature points in the number of the feature points of the user, and determine whether the initial liveness recognition probability of the user does not meet the requirements based on the comprehensive deviation amount of the deviation feature points of the user, if so, determine that the initial liveness recognition probability of the user does not meet the requirements, if not, proceed to the next step;

[0054] The feature points of the user whose depth information credibility is greater than the preset credibility are taken as credible feature points, and the initial liveness recognition probability of the user is determined based on the number of the user's credible feature points, the depth information credibility of different credible feature points, the comprehensive deviation of the user's deviation feature points, and the number of feature points.

[0055] S3: acquiring the infrared image of the user's face, and determining that the infrared living body recognition probability of the user meets the requirement based on the deviation between the temperature characteristics of different characteristic areas of the infrared image of the face and the historical temperature characteristics of different characteristic areas of the user, and then proceeding to the next step;

[0056] In a possible embodiment, the method for determining in the above step S3 that the infrared living body recognition probability of the user meets the requirement is:

[0057] S31 determines the preset temperature intervals of the different characteristic areas of the user and the temperature deviation intervals between the different characteristic areas according to the historical temperature characteristics of the different characteristic areas of the user, and determines the temperatures of the different characteristic areas according to the temperature characteristics of the different characteristic areas of the user, and determines whether the user has a characteristic area whose temperature is not in the preset temperature interval. If so, proceed to step S33, if not, proceed to the next step;

[0058] S32 determines whether there is a characteristic area whose temperature deviation is not within the temperature deviation range based on the temperatures of the different characteristic areas, and if so, proceeds to the next step; if not, determines that the infrared living body recognition probability of the user meets the requirement;

[0059] S33: taking the characteristic area whose temperature is not in the preset temperature interval as the deviation temperature characteristic area, and determining the temperature deviation of the user according to the number of the deviation temperature characteristic areas of the user and the temperature deviations of the temperatures of different deviation temperature areas and the end points of the preset temperature interval, and determining whether the infrared liveness recognition probability of the user does not meet the requirement based on the temperature deviation of the user, if so, determining that the infrared liveness recognition probability of the user does not meet the requirement, if not, proceeding to the next step;

[0060] S34: determining the screening feature areas whose temperature deviations are not within the temperature deviation interval based on the temperatures of the different feature areas and the temperature deviation intervals between the different feature areas, and determining the comparative temperature deviation amount of the user based on the number of the screening feature areas and the deviation amount between the temperature deviations between the screening feature areas and the end points of the temperature deviation interval, and determining whether the infrared liveness recognition probability of the user does not meet the requirement based on the comparative temperature deviation amount of the user; if so, determining that the infrared liveness recognition probability of the user does not meet the requirement; if not, proceeding to the next step;

[0061] S35 obtains the number of characteristic areas where the user's temperature is within the preset temperature range and the number of temperature deviations between the characteristic areas within the temperature deviation range, and determines the infrared liveness recognition probability of the user and whether the infrared liveness recognition probability meets the requirements in combination with the user's comparative temperature deviation and the temperature deviation.

[0062] In another possible embodiment, the method for determining in the above step S3 that the infrared living body recognition probability of the user meets the requirement is:

[0063] Determining preset temperature intervals of different characteristic areas of the user and temperature deviation intervals between different characteristic areas according to historical temperature characteristics of different characteristic areas of the user;

[0064] The characteristic area whose temperature is not in the preset temperature interval is taken as a deviation temperature characteristic area, and the temperature deviation amount of the user is determined by the number of the deviation temperature characteristic areas of the user and the temperature deviation amount between the temperatures of different deviation temperature areas and the end points of the preset temperature interval, and whether the infrared liveness recognition probability of the user does not meet the requirement is determined based on the temperature deviation amount of the user, if so, it is determined that the infrared liveness recognition probability of the user does not meet the requirement, if not, proceeding to the next step;

[0065] Determine the screening feature area whose temperature deviation between the feature areas is not within the temperature deviation interval based on the temperatures of the different feature areas and the temperature deviation interval between the different feature areas, and determine the comparative temperature deviation amount of the user based on the number of the screening feature areas and the deviation amount between the temperature deviation between the screening feature areas and the end point of the temperature deviation interval, and determine whether the infrared liveness recognition probability of the user does not meet the requirement based on the comparative temperature deviation amount of the user, if so, determine that the infrared liveness recognition probability of the user does not meet the requirement, if not, proceed to the next step;

[0066] The number of characteristic areas where the user's temperature is within the preset temperature range and the temperature deviation between the characteristic areas are obtained, and the infrared liveness recognition probability of the user and whether the infrared liveness recognition probability meets the requirements are determined in combination with the user's contrast temperature deviation and the temperature deviation.

[0067] S4 determines the number of identity authentication failures of the user within a preset time based on the identity information of the user, and determines the user's liveness recognition probability and liveness recognition result through the depth information credibility of different feature points with different identity authentication failure times and deviation feature points.

[0068] In a possible embodiment, the method for determining the liveness recognition probability of the user in the above step S4 is:

[0069] The user's liveness recognition probability and liveness recognition result are determined by the depth information credibility of different feature points with different identity authentication failure times and the deviation feature points.

[0070] Determine the number of identity authentication failures for different feature points to be identified as deviation feature points through deviation feature points of different identity authentication failure times, and use it as the number of deviation identification times of feature points, and determine whether there are feature points whose deviation identification times do not meet the requirements based on the number of identity authentication failures. If so, determine the liveness recognition probability of the user through the number of feature points whose deviation identification times do not meet the requirements. If not, proceed to the next step;

[0071] The feature points identified as deviation feature points are used as screening deviation feature points, and the correction credibility of the screening deviation feature points is determined by the depth information credibility of the screening deviation feature points at different times of identity authentication failure and the number of times they are identified as deviation feature points, and the comprehensive correction credibility of the screening deviation feature points of the user is determined by the number of the screening deviation feature points of the user and the correction credibility of different screening deviation feature points;

[0072] Determine whether the comprehensive correction credibility of the screening deviation feature points of the user meets the requirements, if yes, proceed to the next step, if not, determine the liveness recognition probability of the user through the comprehensive correction credibility of the screening deviation feature points;

[0073] Based on the depth information credibility of different feature points at different times of identity authentication failure, the comprehensive credibility of depth information of different feature points is determined, and the liveness recognition probability of the user is determined by removing the number of feature points of the screening deviation feature points and the comprehensive corrected credibility of the depth information and the screening deviation feature points.

[0074] It should be noted that the liveness recognition result of the user includes liveness and non-liveness.

[0075] Example 2

[0076] On the other hand, the present invention provides a computer system, comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that: the processor executes the above-mentioned liveness detection method when running the computer program.

[0077] Example 3

[0078] On the other hand, Figure 3 As shown, the present invention provides a computer storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned living body detection method.

[0079] Through the above embodiments, the present invention achieves the following beneficial effects:

[0080] 1. The user's initial liveness recognition probability is determined by the credibility of the depth information of different feature points and the deviation feature points, thereby avoiding the technical problem of inaccurate liveness recognition results caused by simply using the deviation of the depth information. By comparing with the preset depth information, it is further combined with the user's identity information to improve the accuracy of liveness recognition.

[0081] 2. The user's infrared liveness recognition probability is determined based on the deviation between the temperature characteristics of different characteristic areas of the facial infrared image and the historical temperature characteristics of different characteristic areas of the user, and then further compared with the user's identity information from the perspective of temperature characteristics, thereby achieving accurate recognition of disguised users such as 3D face masks and ensuring the accuracy of liveness recognition results.

[0082] 3. The user's liveness recognition probability and liveness recognition result are determined by the depth information credibility of different feature points with different numbers of identity authentication failures and the deviation feature points, thereby avoiding the influence of the angle deviation of the user's facial image on the accuracy of the liveness recognition result in a single identity authentication, and further improving the accuracy of the liveness recognition result.

[0083] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0084] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.

Claims

1. A method for detecting a living body, characterized in that: Specifically include: Determining identity information of a user, and determining preset depth information between different feature points of the user and different reference points according to the identity information; Determine the depth information between different feature points of the user's facial image and different reference points through a binocular camera device, and determine the depth information credibility of different feature points of the user and the deviation feature points in combination with the preset depth information, determine the initial liveness recognition probability of the user through the depth information credibility of the different feature points and the deviation feature points, and proceed to the next step when the initial liveness recognition probability does not meet the requirement; Acquire the user's facial infrared image, and determine that the infrared living body recognition probability of the user meets the requirement based on the deviation between the temperature characteristics of different characteristic areas of the facial infrared image and the historical temperature characteristics of different characteristic areas of the user, and then proceed to the next step; The number of identity authentication failures of the user within a preset time is determined based on the identity information of the user, and the liveness recognition probability and liveness recognition result of the user are determined through the depth information credibility of different feature points with different numbers of identity authentication failures and deviation feature points.

2. The method for detecting a living body according to claim 1, wherein: The identity information of the user is determined based on the recognition result of the user's facial image, and specifically based on the matching result between the user's facial image and a facial image database.

3. The method for detecting a living body according to claim 1, wherein: The preset depth information is determined according to the depth information of the reserved facial image of the user in the facial image database or according to the depth information of the facial image of the user during historical identity authentication.

4. The method for detecting a living body according to claim 1, wherein: The method for determining the credibility of the depth information of the feature point is: Determine the deviation of the depth information between the feature point and different reference points through the depth information between the feature point and different reference points and the difference of the preset depth information, and judge whether there is a reference point whose deviation does not meet the requirement, if yes, proceed to the next step, if not, determine the credibility of the depth information of the feature point through the average value of the deviation of the depth information between the feature point and different reference points; The reference point whose deviation meets the requirement is taken as a credible reference point, and whether the reference point is credible is determined according to the number of the credible reference points and their proportion in the reference points. If so, the credibility of the depth information of the feature point is determined by the average value of the deviation of the depth information between the feature point and different reference points. If not, proceed to the next step; The reference point whose deviation does not meet the requirement is taken as the deviation reference point, and the comprehensive deviation of the deviation reference point is determined according to the number of the deviation reference points, the maximum value and the average value of the deviation of the deviation reference point, and it is judged whether the comprehensive deviation of the deviation reference point meets the requirement, if so, proceed to the next step, if not, determine the credibility of the depth information of the feature point through the comprehensive deviation; The credibility of the depth information of the feature point is determined based on the number of the credible reference points, the average value of the deviation of the depth information between the feature point and different reference points, and the comprehensive deviation.

5. The method for detecting a living body according to claim 1, wherein: When the credibility of the depth information of the feature point does not meet the requirement, the feature point is determined to be a deviation feature point.

6. The method for detecting a living body according to claim 1, wherein: The method for determining whether the infrared living body recognition probability of the user meets the requirements is: Determining preset temperature intervals of different characteristic areas of the user and temperature deviation intervals between different characteristic areas according to historical temperature characteristics of different characteristic areas of the user; The characteristic area whose temperature is not in the preset temperature interval is taken as a deviation temperature characteristic area, and the temperature deviation amount of the user is determined by the number of the deviation temperature characteristic areas of the user and the temperature deviation amount between the temperatures of different deviation temperature areas and the end points of the preset temperature interval, and whether the infrared liveness recognition probability of the user does not meet the requirement is determined based on the temperature deviation amount of the user, if so, it is determined that the infrared liveness recognition probability of the user does not meet the requirement, if not, proceeding to the next step; Determine the screening feature area whose temperature deviation between the feature areas is not within the temperature deviation interval based on the temperatures of the different feature areas and the temperature deviation interval between the different feature areas, and determine the comparative temperature deviation amount of the user based on the number of the screening feature areas and the deviation amount between the temperature deviation between the screening feature areas and the end point of the temperature deviation interval, and determine whether the infrared liveness recognition probability of the user does not meet the requirement based on the comparative temperature deviation amount of the user, if so, determine that the infrared liveness recognition probability of the user does not meet the requirement, if not, proceed to the next step; The number of characteristic areas where the user's temperature is within the preset temperature range and the temperature deviation between the characteristic areas are obtained, and the infrared liveness recognition probability of the user and whether the infrared liveness recognition probability meets the requirements are determined in combination with the user's contrast temperature deviation and the temperature deviation.

7. The method for detecting a living body according to claim 1, wherein: The method for determining the user's liveness recognition probability is: Determine the number of identity authentication failures for different feature points to be identified as deviation feature points through deviation feature points of different identity authentication failure times, and use it as the number of deviation identification times of feature points, and determine whether there are feature points whose deviation identification times do not meet the requirements based on the number of identity authentication failures. If so, determine the liveness recognition probability of the user through the number of feature points whose deviation identification times do not meet the requirements. If not, proceed to the next step; The feature points identified as deviation feature points are used as screening deviation feature points, and the correction credibility of the screening deviation feature points is determined by the depth information credibility of the screening deviation feature points at different times of identity authentication failure and the number of times they are identified as deviation feature points, and the comprehensive correction credibility of the screening deviation feature points of the user is determined by the number of the screening deviation feature points of the user and the correction credibility of different screening deviation feature points; Determine whether the comprehensive correction credibility of the screening deviation feature points of the user meets the requirements, if yes, proceed to the next step, if not, determine the liveness recognition probability of the user through the comprehensive correction credibility of the screening deviation feature points; Based on the depth information credibility of different feature points at different times of identity authentication failure, the comprehensive credibility of depth information of different feature points is determined, and the liveness recognition probability of the user is determined by removing the number of feature points of the screening deviation feature points and the comprehensive corrected credibility of the depth information and the screening deviation feature points.

8. The method for detecting a living body according to claim 1, wherein: The liveness recognition result of the user includes liveness and non-liveness.

9. A computer system comprising: A memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes a liveness detection method according to any one of claims 1 to 8 when running the computer program.

10. A computer storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute a liveness detection method according to any one of claims 1 to 8.

Images