Identity authentication method, device, electronic device and storage medium
By introducing facial feature library and biometric comparison of similar objects in facial recognition technology, the problem of low identity authentication efficiency in high similarity scenarios is solved, and fast and accurate identity authentication is achieved.
Patent Information
- Application Number
- CN202110847682.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-07-27
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-07-27
AI Technical Summary
The existing facial recognition technology has low recognition efficiency in high similarity scenarios, resulting in the extension of the identity authentication process, especially when the facial feature database data is large, the feature retrieval and comparison time is long, which affects the identity authentication efficiency.
By obtaining the target face features of the object to be authenticated and preferentially searching the facial feature library of similar objects. If the similarity is greater than the threshold, the target site biometrics are obtained for similarity comparison to speed up the identity authentication process.
On the basis of ensuring the accuracy of identity authentication, the efficiency of identity authentication is improved, the time to identify high-similar users is reduced, and the speed of identity authentication is improved.
Smart Images

Figure CN115700835B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to an identity authentication method, device, electronic equipment and storage medium. Background Art
[0002] With the rapid development of the current economy, facial recognition technology has been widely used in various business scenarios that require very high accuracy, such as common online payments, community or company access control systems, etc. These scenarios require that the identity recognition process cannot be wrong, otherwise it will cause serious consequences.
[0003] At present, in order to ensure the accuracy of face recognition, in the process of face recognition, the facial features are generally compared with the facial features of all users in the database to obtain the result of identity recognition. With this solution, since the facial feature database usually has a large amount of data, the feature retrieval and comparison process will take longer and the comparison speed will be slow, which affects the efficiency of identity authentication. Summary of the invention
[0004] The embodiments of the present invention provide an identity authentication method, device, electronic device and storage medium, which can accelerate the process of feature retrieval and comparison and improve the efficiency of identity authentication for users whose identities cannot be accurately authenticated only through facial recognition, while ensuring the accuracy of identity authentication.
[0005] An embodiment of the present invention provides an identity authentication method, including:
[0006] Acquire the target facial features of the object to be authenticated and read a similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each of the facial feature groups includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold;
[0007] Comparing the target facial features with the facial features of similar objects in the similar object facial feature library in terms of facial similarity;
[0008] If the similarity between the target facial feature and at least one of the similar object facial features is greater than the first facial similarity threshold, acquiring the target part biometric feature of the object to be authenticated;
[0009] Comparing the target part biometric features of the object to be authenticated with the target part biometric features of each of the similar objects in the similar object facial feature library for similarity;
[0010] Based on the similarity comparison result of the biological features of the target part, the identity of the object to be authenticated is authenticated.
[0011] Accordingly, an embodiment of the present invention provides an identity authentication device, including:
[0012] A facial feature acquisition unit, used to acquire a target facial feature of an object to be authenticated and read a similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each of the facial feature groups includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold;
[0013] A feature comparison unit, used for comparing the target facial features with the facial features of similar objects in the similar object facial feature library in terms of facial similarity;
[0014] a biometric feature acquisition unit, configured to acquire a target part biometric feature of the object to be authenticated if the similarity between the target facial feature and at least one facial feature of the similar object is greater than the first facial similarity threshold;
[0015] A biometric feature comparison unit, used for comparing the biometric features of the target part of the object to be authenticated with the biometric features of the target part of each of the similar objects in the similar object facial feature library for similarity;
[0016] The identity authentication unit is used to perform identity authentication on the object to be authenticated based on the similarity comparison result of the biological characteristics of the target part.
[0017] Optionally, the identity authentication device further includes:
[0018] a global feature comparison unit, configured to compare the target facial feature with global facial features in a global facial feature library for facial similarity if the similarity between the target facial feature and the facial features of each of the similar objects is not greater than the first facial similarity threshold, wherein the global facial feature library includes global facial features of multiple global objects;
[0019] The second identity authentication unit is used to perform identity authentication on the object to be authenticated based on the result of the facial similarity comparison.
[0020] Optionally, the second identity authentication unit is used to obtain the target part biometric feature of the object to be authenticated if the similarity between the target facial feature and at least two of the global facial features is greater than a second facial similarity threshold;
[0021] Based on the target part biometric features of the object to be authenticated and the target part biometric features of each of the global objects in the global facial feature library, performing a similarity comparison of the target part biometric features;
[0022] Based on the similarity comparison result of the biological features of the target part, the identity of the object to be authenticated is authenticated.
[0023] Optionally, the second identity authentication unit is used to determine the target part biometric feature of the target global object from the target part biometric features of each of the global objects in the global facial feature library, and the similarity between the global facial feature of the target global object and the target facial feature is greater than the second facial similarity threshold;
[0024] The target part biometric feature of the object to be authenticated is compared with the target part biometric features of each target global object for similarity.
[0025] Optionally, the identity authentication device further includes:
[0026] A global object identity authentication unit is used to authenticate the object to be authenticated based on global object information corresponding to the target number of global facial features if the similarity between the target facial feature and the target number of global facial features in the global facial feature library is greater than the second facial similarity threshold.
[0027] Optionally, before the global feature comparison unit, it further includes:
[0028] A sample feature comparison unit, used to obtain a sample facial feature library, wherein the sample objects in the sample facial feature library have undergone identity authentication at least once;
[0029] Comparing the target facial features with the sample facial features in the sample facial feature library for facial similarity;
[0030] If the similarity between the target facial feature and each of the sample facial features is not greater than a preset third facial similarity threshold, the step of comparing the facial similarity between the target facial feature and the global facial features in the global facial feature library is performed.
[0031] Optionally, the sample feature comparison unit further includes: if the similarity between the target facial feature and a preset number of sample facial features is greater than a preset third facial similarity threshold, performing identity authentication on the object to be authenticated based on sample object information corresponding to the preset number of sample facial features.
[0032] Optionally, the sample facial feature library includes a first access queue, and the first access queue can store a certain number of sample facial features. The second identity authentication unit also includes a first feature storage unit, which is used to add the target object identifier of the object to be authenticated and the target facial feature of the object to be authenticated to the first access queue accordingly, and delete the object identifier of the sample object with the longest storage time in the first access queue and the corresponding sample facial feature.
[0033] Optionally, the first feature storage unit also includes a second feature storage unit, which is used to add the target object identifier and the facial features of the object to be authenticated to the second access queue if there is an object identifier identical to the target object identifier in the first access queue, and delete the target object identifier and the corresponding target facial features from the first access queue;
[0034] If there is an object identifier identical to the target object identifier in the second access queue, the target object identifier is moved to the head of the second access queue.
[0035] Optionally, the similar object facial feature library corresponds to an authentication service, and the facial feature acquisition unit further includes a feature library determination unit, which is used to obtain an identity authentication request, the identity authentication request includes authentication service information, and the authentication service information indicates the authentication service;
[0036] Based on the authentication service information, a facial feature library of similar objects corresponding to the authentication service is determined.
[0037] Optionally, the identity authentication device provided by the embodiment of the present invention further includes a registration unit, which is used to obtain the registered facial features of the registered object, and compare the similarity between the registered facial features and each of the global facial features in the global facial feature library;
[0038] If there is at least one of the global facial features and the registered facial feature whose similarity is greater than the first facial similarity threshold, the at least one global facial feature and the registered facial feature are added to the similar object facial feature library.
[0039] Optionally, the facial feature acquisition unit further includes an identification acquisition unit, which is used to acquire a similar identification of the object to be authenticated, wherein the similar identification indicates whether the object to be authenticated is a similar object;
[0040] When the similarity identifier indicates that the object to be authenticated is a similar object, the step of acquiring the target facial features of the object to be authenticated is not performed, and the step of acquiring the target part biometric features of the object to be authenticated is performed.
[0041] Correspondingly, an embodiment of the present invention further provides an electronic device, comprising a memory and a processor; the memory stores an application, and the processor is used to run the application in the memory to execute the steps in any one of the identity authentication methods provided in the embodiments of the present invention.
[0042] In addition, an embodiment of the present invention further provides a computer-readable storage medium, which stores a plurality of instructions, and the instructions are suitable for a processor to load to execute the steps in any one of the identity authentication methods provided in the embodiment of the present invention.
[0043] By adopting the scheme of the embodiment of the present invention, the target facial features of the object to be authenticated can be obtained and the similar object facial feature library can be read. The similar object facial feature library stores at least one facial feature group, each of which includes similar object facial features of at least two similar objects. The facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold. The target facial features are compared with the facial features of the similar objects in the similar object facial feature library for facial similarity. If the similarity between the target facial features and at least one of the similar object facial features is greater than the first facial similarity threshold, the target part biometric features of the object to be authenticated are obtained, and the target part biometric features of the object to be authenticated are compared. The similarity between the feature and the target part biometric feature of each similar object in the facial feature library of the similar object is compared, and the identity of the object to be authenticated is authenticated based on the similarity comparison result of the target part biometric feature; since after obtaining the target facial feature of the object to be authenticated, the facial feature library of the similar object is retrieved first, and the size of the facial feature library of the similar object is much smaller than the facial feature library that stores the facial features of all objects, it can quickly determine whether the object to be authenticated is a similar object, which accelerates the identity recognition process of highly similar users belonging to similar objects, and enters the similarity comparison stage of the target part biometric feature as soon as possible. Therefore, on the basis of ensuring the accuracy of identity authentication, the efficiency of identity authentication of the object to be authenticated is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0045] Figure 1 is a schematic diagram of a scenario of an identity authentication method provided by an embodiment of the present invention;
[0046] Figure 2 is a flow chart of an identity authentication method provided by an embodiment of the present invention;
[0047] Figure 3 It is a schematic diagram of a feature retrieval process in an identity authentication process provided by an embodiment of the present invention;
[0048] Figure 4 is another flow chart of the identity authentication method provided by an embodiment of the present invention;
[0049] Figure 5 It is a schematic diagram of the architecture of the identity authentication system provided by an embodiment of the present invention;
[0050] Figure 6 is a schematic diagram of a first access queue and a second access queue provided by an embodiment of the present invention;
[0051] Figure 7 is a schematic diagram of the structure of an identity authentication device provided by an embodiment of the present invention;
[0052] Figure 8 is another structural schematic diagram of an identity authentication device provided by an embodiment of the present invention;
[0053] Fig. 9 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0055] The embodiment of the present invention provides an identity authentication method, device, electronic device and computer-readable storage medium. Specifically, the embodiment of the present invention provides an identity authentication method applicable to an identity authentication device, and the identity authentication device can be integrated in an electronic device.
[0056] The electronic device can be a terminal or other device, including but not limited to a mobile terminal and a fixed terminal. For example, a mobile terminal includes but is not limited to a smart phone, a smart watch, a tablet computer, a laptop computer, a smart car-mounted terminal, etc., wherein a fixed terminal includes but is not limited to a desktop computer, a smart TV, etc.
[0057] The electronic device can also be a server or other device. The server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, as well as big data and artificial intelligence platforms, but is not limited to these.
[0058] The identity authentication method of the embodiment of the present invention may be implemented by a terminal, or may be implemented by a terminal and a server together.
[0059] The following takes the example of a terminal and a server jointly implementing the identity authentication method to illustrate the method.
[0060] like Figure 1 As shown, the identity authentication system provided by the embodiment of the present invention includes a terminal 10 and a server 20, etc.; the terminal 10 and the server 20 are connected via a network, for example, via a wired or wireless network connection, etc., wherein the terminal 10 can exist as a terminal for obtaining the target facial features of the object to be authenticated.
[0061] The terminal 10 may be a terminal for collecting target facial features of the object to be authenticated, and is used to send the facial features that need to be compared for similarity to the server 20 .
[0062] The server 20 can obtain the target facial features of the object to be authenticated and read the similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each facial feature group includes similar object facial features of at least two similar objects, the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold, and the target facial features are compared with the facial features of the similar objects in the similar object facial feature library for facial similarity.
[0063] If the similarity between the target facial features and the facial features of at least one similar object is greater than the first facial similarity threshold, the terminal 10 can be triggered to collect the target part biometric features of the object to be authenticated, and send the target part biometric features of the object to be authenticated that need to be compared for similarity to the server 20.
[0064] The server 20 may compare the target part biometric features of the object to be authenticated with the target part biometric features of each similar object in the similar object facial feature library, and perform identity authentication on the object to be authenticated based on the similarity comparison result of the target part biometric features.
[0065] In other embodiments, the identity authentication method can be implemented by a terminal. The identity authentication system provided by the embodiment of the present invention includes a terminal, which stores a similar object facial feature library and target part biometric features of each similar object in the similar object facial feature library.
[0066] The terminal can obtain the target facial features of the object to be authenticated and read the similar object facial feature library, wherein at least one facial feature group is stored in the similar object facial feature library, each facial feature group includes similar object facial features of at least two similar objects, the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold, and the facial similarity of the target facial features is compared with the facial features of the similar objects in the similar object facial feature library. If the similarity between the target facial features and at least one similar object facial feature is greater than the first facial similarity threshold, the target part biometric feature of the object to be authenticated is obtained, and the target part biometric feature of the object to be authenticated is compared with the target part biometric features of each similar object in the similar object facial feature library for similarity, and based on the similarity comparison result of the target part biometric feature, the identity of the object to be authenticated is authenticated.
[0067] It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.
[0068] The embodiments of the present invention will be described from the perspective of an identity authentication device, which may be specifically integrated in a server and / or a terminal.
[0069] like Figure 2 As shown, the specific process of the identity authentication method of this embodiment can be as follows:
[0070] 201. Obtain target facial features of an object to be authenticated and read a similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each facial feature group includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold.
[0071] The object to be authenticated is an object for identity authentication, and the object for identity authentication may be a human being or other animals, etc., which is not limited in the embodiment of the present invention.
[0072] It is understandable that the object to be authenticated each time the identity authentication is performed may be one, or two or more. For example, the facial features of multiple users may be simultaneously collected by a facial feature acquisition device, and the identity authentication device may simultaneously implement identity authentication of two or more users through multi-threaded parallelism or the like.
[0073] Among them, the process of obtaining the target facial features of the object to be authenticated generally requires combining computer vision technology and machine learning technology in artificial intelligence to obtain the user's facial features.
[0074] Computer Vision Technology (CV) Computer vision is a science that studies how to make machines "see". To put it more specifically, it refers to machine vision such as using cameras and computers to replace human eyes to identify and measure targets, and further perform graphic processing to make computer processing into images that are more suitable for human eye observation or transmission to instruments for detection. As a scientific discipline, computer vision studies related theories and technologies, and attempts to establish an artificial intelligence system that can obtain information from images or multi-dimensional data. Computer vision technology usually includes image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous positioning and map construction, and other technologies, as well as common biometric recognition technologies such as face recognition and fingerprint recognition.
[0075] For example, the step of "obtaining facial features of the object to be authenticated" may specifically include: after the image acquisition device acquires the facial image of the object to be authenticated, facial features of the facial image of the object to be authenticated may be extracted by using a neural network or a facial feature extraction model that can extract facial features to obtain the facial features of the object to be authenticated.
[0076] Alternatively, the step of "obtaining the target facial features of the object to be authenticated" may specifically include: collecting an image containing the face of the object to be authenticated, using a key point positioning method to perform vector representation on the face of the object to be authenticated in the image, and using the obtained vector as the facial features of the object to be authenticated. Specifically, the key point positioning method may include active shape models (ASM), deep convolutional networks (DCNN), deep alignment networks (DAN), and other methods.
[0077] Alternatively, the step of "obtaining facial features of the object to be authenticated" may specifically include: collecting an image containing the face of the object to be authenticated, obtaining the facial image of the object to be authenticated therefrom, performing image segmentation on the facial image through image processing technology in computer vision technology, for example, segmenting the facial image into sub-images containing only eyes, only lips, only nose and other facial feature parts, then performing vector representation on each sub-image, and using the vector corresponding to each sub-image as the facial feature of the object to be authenticated.
[0078] In some optional embodiments, in order to further improve the efficiency of identity authentication, certain identifiers may be set for objects belonging to similar objects to distinguish objects belonging to similar objects from other objects. When it is determined based on the identifier that the object logged in on the terminal belongs to a similar object, the face recognition process may be omitted and the comparison process of the biometric features of the target part may be performed directly. Therefore, before step 201, the following may also be included:
[0079] Obtaining a similarity identifier of the object to be authenticated, the similarity identifier indicating whether the object to be authenticated is a similar object;
[0080] When the similarity identifier indicates that the object to be authenticated is a similar object, the step of acquiring the target facial features of the object to be authenticated is not performed, and the step of acquiring the target part biometric features of the object to be authenticated is performed.
[0081] Among them, the similarity identifier may include an identifier indicating that the object to be authenticated is a similar object and an identifier indicating that the object to be authenticated is not a similar object. For example, the similarity identifier may include 0 and 1. If the similarity identifier of the object to be authenticated is 0, the object to be authenticated is a similar object, and so on.
[0082] 202. Compare the target facial features with the facial features of similar objects in the facial feature library of similar objects in terms of facial similarity.
[0083] Among them, the facial similarity comparison process applies the facial recognition technology. Facial recognition is a biometric technology that identifies an individual based on facial feature information. A series of related technologies that use image acquisition devices such as cameras or webcams to collect images or video streams containing faces, automatically detect faces in images, and then perform facial recognition on the detected faces are usually also called facial recognition.
[0084] In the face recognition process, usually one or a group of face images can be represented as a high-dimensional vector after extracting key facial features through a deep model. An input high-dimensional vector (target feature vector) is compared with multiple known feature vectors in the feature library (similar object face feature library, sample face feature library and / or global face feature library) one by one to find the most similar one or more feature vectors. A common calculation method is to use the cosine distance to represent the distance between two vectors, which can also be understood as feature similarity.
[0085] In some embodiments, the facial similarity between the target facial features and the facial features of similar objects in the similar object facial feature library can be obtained by calculating the Euclidean distance, Pearson correlation coefficient, cosine similarity, etc. between the target feature vector and one or more known feature vectors.
[0086] 203. If the similarity between the target facial feature and the facial feature of at least one similar object is greater than a first facial similarity threshold, obtain a target part biometric feature of the object to be authenticated.
[0087] The first facial similarity threshold may be a value preset by a technician or an administrator of the identity authentication system. For example, the first facial similarity threshold may be a value such as 99% or 0.99.
[0088] It is understandable that the current facial recognition algorithms cannot achieve 100% accurate recognition, for example, when dealing with the identification of highly similar people, such as identical twins, relatives with blood relations, and highly similar people with "common faces" whose facial features are very similar to those of other objects, etc. Or, in order to ensure health, more and more people wear masks. Although mask wearing recognition algorithms have been technically developed, since wearing masks blocks most of the facial information, the recognition accuracy will be lower than when not wearing masks.
[0089] However, in business scenarios that require high accuracy of identity authentication, such as face payment, community or company access control systems, etc., if the result of identity authentication is wrong, it will cause great inconvenience to business users. In actual application, when the similarity between the target facial features of the object to be authenticated and the facial features of at least one similar object is greater than the first facial similarity threshold, in order to ensure the accuracy of identity authentication, it can be considered that the identity of the object to be authenticated cannot be determined by facial recognition alone.
[0090] For example, a pair of twins A and B, when A is used as the object to be authenticated, at least two similar object facial features (A and B's facial features) will be matched in the similar object facial feature library. The similarity between these two similar object facial features and A's target facial features is greater than the first facial similarity threshold. At this time, A's identity cannot be determined.
[0091] Therefore, for an object to be authenticated that cannot be authenticated only through the facial recognition process, identity authentication can be performed in combination with biometric features of target parts other than the face.
[0092] Among them, the target part biometric feature is the biometric feature of the preset part, for example, the preset part can be the eye, hand, finger part, palm part, etc. Therefore, the target part biometric feature can be iris feature, retina feature, hand shape feature, fingerprint feature, palm print feature, etc.
[0093] For example, palm print recognition is a contactless recognition method that is easy for users of the identity authentication system to accept, and the collection of palm prints does not require high requirements for collection equipment. Compared with requiring users to enter private information such as mobile phone numbers or choose other payment methods that do not use facial recognition, it is more user-friendly and greatly reduces the complexity of the operation, making it easier for users to accept.
[0094] In some optional embodiments, the step of “obtaining biometric features of a target part of the object to be authenticated” may include: obtaining palm print features of the object to be authenticated.
[0095] The shape of palm prints is controlled by genetic genes. Even if the epidermis peels off for some reason, the new palm prints still maintain the original structure. Everyone's palm prints are different. Even for twins, their palm prints are only similar, not exactly the same. Therefore, it can be used to make up for the problem of high similarity of individual objects faced by face recognition.
[0096] 204. Compare the target part biometric features of the object to be authenticated with the target part biometric features of each similar object in the similar object facial feature database for similarity.
[0097] Among them, usually one or a group of target part images can be represented as a high-dimensional vector after extracting key target part biometric features through a deep model. An input high-dimensional vector (target part biometric features of the object to be authenticated) is compared with multiple known feature vectors in the feature library (target part biometric features of each similar object in the similar object facial feature library) one by one to find the most similar feature vector.
[0098] In some examples, if no facial features of a similar facial object whose similarity with the target facial features is greater than a first facial similarity threshold are matched in the similar object facial feature library, the target facial features may be compared with the facial features of all objects in the identity authentication system for similarity to determine the identity of the object to be authenticated.
[0099] That is, before step 204 or the identity authentication method provided by the embodiment of the present invention may further include:
[0100] If the similarity between the target facial feature and the facial features of each similar object is not greater than a first facial similarity threshold, comparing the target facial feature with global facial features in a global facial feature library for facial similarity, where the global facial feature library includes global facial features of multiple global objects;
[0101] Based on the result of the facial similarity comparison, the identity of the object to be authenticated is authenticated.
[0102] The global facial feature library stores the facial features of all objects that have used the identity authentication system. In actual application, as more and more objects use the identity authentication system, the number of global facial features in the global facial feature library may also increase continuously.
[0103] Optionally, if a subject uses the identity authentication system multiple times, the global facial features corresponding to the subject in the global facial feature library can be updated according to the target facial features of the subject each time the identity authentication is performed. The specific updating method can be to replace the existing global facial features of the subject in the global facial feature library with the new target facial features to obtain new global facial features corresponding to the subject; or, each time a new target facial feature is obtained, the new target facial features are fused with the existing global facial features of the subject in the global facial feature library to obtain new global facial features corresponding to the subject, etc. Feature fusion can be achieved by means of convolutional networks, weighted calculations, etc.
[0104] Specifically, the step of “authenticating the identity of the object to be authenticated based on the result of the facial similarity comparison” may include:
[0105] If the similarity between the target facial feature and the at least two global facial features is greater than a second facial similarity threshold, obtaining a target part biometric feature of the object to be authenticated;
[0106] Based on the target part biometric features of the object to be authenticated and the target part biometric features of each global object in the global facial feature library, a similarity comparison is performed on the target part biometric features;
[0107] Based on the similarity comparison result of the biometric features of the target part, the identity of the object to be authenticated is authenticated.
[0108] When the similarity comparison of the target part biometric features is performed, the similarity of the target part biometric features of each global object in the global facial feature library and the target part biometric features of the object to be authenticated may be calculated respectively.
[0109] Alternatively, the global objects corresponding to the global facial features whose similarity with the target facial features is greater than the second facial similarity threshold may be determined, and only the target part biometric features of these global objects are obtained for comparison with the target part biometric features of the object to be authenticated. In other words, the step of "comparing the similarity of the target part biometric features based on the target part biometric features of the object to be authenticated and the target part biometric features of each global object in the global facial feature library" may include:
[0110] Determining the target part biometric features of the target global object from the target part biometric features of each global object in the global facial feature library, wherein the similarity between the global facial features of the target global object and the target facial features is greater than a second facial similarity threshold;
[0111] The target part biometric features of the object to be authenticated are compared with the target part biometric features of each target global object for similarity.
[0112] The second facial similarity threshold may be a value preset by a technician or an administrator of the identity authentication system. For example, the second facial similarity threshold may be a value such as 98% or 0.98. The second facial similarity threshold may be the same as the first facial similarity threshold, or may be different from the first facial similarity threshold, which is not limited in the embodiment of the present invention.
[0113] The target global object is a global object corresponding to a global facial feature whose similarity with the target facial feature is greater than a second facial similarity threshold.
[0114] In some optional embodiments, the step of “authenticating the identity of the subject to be authenticated based on the result of the facial similarity comparison” may include:
[0115] If the similarity between the target facial feature and a target number of global facial features in the global facial feature library is greater than a second facial similarity threshold, the identity of the object to be authenticated is performed based on the global object information corresponding to the target number of global facial features.
[0116] Among them, the global object information is information related to the global object, for example, the global object information may include authentication behavior information such as the object historical authentication time and historical authentication location of the global object, and may also include global object identity information, such as the object identifier of the global object, the object account number and the corresponding password. In some embodiments, the global object information can be stored corresponding to the global object identity information, and so on. The target number is a value set by the technician or the administrator of the identity authentication system. For example, the target number can be 1, 2, 10, etc., and the present invention does not limit this.
[0117] For example, when the number of targets is 1, the global object information corresponding to the global facial feature can be directly used to determine the global object identity information as the identity information of the object to be authenticated, and identity authentication can be performed.
[0118] For another example, when the number of targets is 2 or more, the global object information corresponding to these global facial features can be obtained, and based on the object's historical authentication time, post-authentication consumption amount, historical authentication location and other information in the global object information, as well as the authentication behavior information of the object to be authenticated, such as the authentication time information, post-authentication consumption amount, historical authentication location and other authentication behavior information, the global object information with the highest similarity to the authentication behavior information of the object to be authenticated can be obtained, and based on the global object information, the global object identity information is obtained as the identity information of the object to be authenticated for identity authentication, and so on.
[0119] Alternatively, the global object information corresponding to the global facial feature with the highest similarity to the target facial feature may be directly obtained, and the identity of the object to be authenticated may be authenticated based on the global object information, and so on.
[0120] In actual applications, if the facial features stored in the facial feature library have been accessed recently, then compared with other facial features that have not been accessed, these facial features that have been accessed are more likely to be accessed in the future. In the identity authentication scenario, as time goes by, a large number of objects may be registered in the identity authentication system. Some objects are high-frequency objects and often use the system for identity authentication and other operations, while some objects are low-frequency objects and rarely access the system. Some objects may even be registered and used only once and then never used again.
[0121] Therefore, when storing facial features, a facial feature library of recently frequently accessed users can be constructed in the storage space to speed up the process of retrieving and comparing the similarity of the facial features of the object during identity recognition.
[0122] That is, before the step of “comparing the target facial features with the global facial features in the global facial feature library for facial similarity”, the following steps may also be included:
[0123] Obtaining a sample facial feature library, wherein the sample objects in the sample facial feature library have undergone identity authentication at least once;
[0124] Compare the target facial features with the sample facial features in the sample facial feature library for facial similarity;
[0125] If the similarity between the target facial feature and each sample facial feature is not greater than a preset third facial similarity threshold, a step of comparing the facial similarity between the target facial feature and the global facial features in the global facial feature library is performed.
[0126] The sample object has undergone an identity authentication process at least once, that is, the sample facial features corresponding to the sample object have been accessed at least once.
[0127] In actual application, in order to avoid too many sample facial features in the sample facial feature library, which affects the speed of facial feature retrieval, the sample objects in the sample facial feature library can be set to be objects that have undergone identity authentication within a recent period of time, such as the last 24 hours, the last week, etc. The specific time period can be set by the technician, and the embodiment of the present invention is not limited to this.
[0128] In some optional examples, the identity authentication method provided by the embodiment of the present invention may further include:
[0129] If the similarity between the target facial feature and a preset number of sample facial features is greater than a preset third facial similarity threshold, the identity of the object to be authenticated is performed based on the sample object information corresponding to the preset number of sample facial features.
[0130] The third facial similarity threshold may be a value preset by a technician or an administrator of the identity authentication system. For example, the third facial similarity threshold may be a value such as 97% or 0.97. The third facial similarity threshold may be the same as the first facial similarity threshold and / or the second facial similarity threshold, or may be different from the first facial similarity threshold and / or the second facial similarity threshold, which is not limited in the embodiments of the present invention.
[0131] Among them, the preset number is a target number and a numerical value set by the technician or the administrator of the identity authentication system. For example, the target number can be 1, 2, 5, etc. The preset number can be the same as the aforementioned target number or different from the aforementioned target number. The present invention does not limit this.
[0132] The sample object information is information related to the sample object, for example, the sample object information may include authentication behavior information such as the historical authentication time and historical authentication location of the sample object, and may also include sample object identity information, such as the object identifier of the sample object, the object account number and the corresponding password, etc. In some embodiments, the sample object information may be stored corresponding to the sample object identity information, and so on.
[0133] For example, when the preset number is 1, the sample object identity information can be directly determined as the identity information of the object to be authenticated based on the sample object information corresponding to the sample facial feature, and identity authentication can be performed.
[0134] For another example, when the preset number is 2 or more, the information of the sample objects corresponding to these sample facial features can be obtained, and based on the authentication behavior information such as the object's historical authentication time, post-authentication consumption amount, historical authentication location, etc. in the sample object information, and the authentication behavior information such as the authentication time information, post-authentication consumption amount, historical authentication location, etc. of the object to be authenticated, the sample object information with the highest similarity to the authentication behavior information of the object to be authenticated is obtained, and the sample object identity information is determined from the sample object information as the identity information of the object to be authenticated for identity authentication, and so on.
[0135] Alternatively, the sample object information corresponding to the sample facial feature with the highest similarity to the target facial feature may be directly obtained, the sample object identity information may be determined according to the sample object information, and the identity of the object to be authenticated may be authenticated, and so on.
[0136] In some optional embodiments, when the preset number is at least two, the step of “performing identity authentication on the object to be authenticated based on the sample object information corresponding to the preset number of sample facial features” includes:
[0137] Obtaining a target part biometric feature of the object to be authenticated, and comparing the target part biometric feature of the object to be authenticated with a preset number of target part biometric features of sample objects in a sample facial feature library for similarity;
[0138] Based on the similarity comparison results of the biometric features of the target part and the sample object information corresponding to a preset number of sample facial features, the identity of the object to be authenticated is authenticated.
[0139] The similarity between the sample facial features of the preset number of sample objects and the target facial features is greater than a third facial similarity threshold.
[0140] In actual application, a queue can be set in the sample facial feature library, which stores the sample facial features and sample object identifiers of the most recently N accessed sample objects to ensure that too many sample facial features are not stored in the sample facial feature library.
[0141] That is, the sample facial feature library includes a first access queue, and the number of sample facial features that can be stored in the first access queue is certain. After the step of "authenticating the object to be authenticated based on the result of the facial similarity comparison", the identity authentication method provided by the embodiment of the present invention may further include:
[0142] The target object identifier of the object to be authenticated and the target facial features of the object to be authenticated are added to the first access queue accordingly, and the object identifier of the sample object with the longest storage time and the corresponding sample facial features in the first access queue are deleted.
[0143] That is, when the facial features of an object are accessed for the first time, such as Figure 3 As shown, the facial feature and the object identifier of the object can be added to the first access queue, wherein the first access queue adopts a first-in-first-out strategy. When the queue reaches the upper capacity limit (the number of sample facial features that can be stored), the sample facial features and sample object identifier of the sample object accessed the longest time ago will be eliminated from the queue.
[0144] Furthermore, in order to prevent the real high-frequency objects in the sample facial feature library from being replaced by a large number of objects that are only accessed once at a certain accidental time, the sample facial feature library may also include a second access queue. Before the step of "adding the target object identifier of the object to be authenticated and the target facial feature of the object to be authenticated to the first access queue in correspondence", the identity authentication method of the embodiment of the present invention may also include:
[0145] If there is an object identifier identical to the target object identifier in the first access queue, the target object identifier and the facial features of the object to be authenticated are added to the second access queue in correspondence, and the target object identifier and the corresponding target facial features are deleted from the first access queue;
[0146] If there is an object identifier identical to the target object identifier in the second access queue, the target object identifier is moved to the head of the second access queue.
[0147] Therefore, when the sample facial features in the first access queue are accessed again, that is, the sample object has been accessed twice recently, the sample facial features and sample object identifier of the sample object will be moved to the second access queue. In the second access queue, the recently added sample facial features and sample object identifier are at the head of the queue, and the data accessed again in the second access queue will also be moved to the head of the queue, ensuring that the user with the higher recent access frequency is less likely to be eliminated from the second access queue. The two-level access queue can improve the hit rate of the sample facial features of the object and avoid the situation where a high-frequency object is replaced by a large number of objects that are only accessed once at a certain accidental time.
[0148] It is understandable that the number of sample facial features that can be stored in the second access queue is also limited. Therefore, when the number of sample facial features stored in the second access queue reaches a preset upper limit, the object identifier of the sample object at the end of the second access queue and the corresponding sample facial features can be deleted. Specifically, the step of adding the target object identifier and the facial features of the object to be authenticated to the second access queue and the step of deleting the object identifier and the corresponding sample facial features of the sample object at the end of the second access queue can be performed simultaneously through multi-threading or the like, or the object identifier and the corresponding sample facial features of the sample object at the end of the second access queue can be deleted first, and then the target object identifier and the facial features of the object to be authenticated can be added to the second access queue, and the present invention does not limit this.
[0149] 205. Based on the similarity comparison result of the biometric features of the target part, the identity of the object to be authenticated is authenticated.
[0150] Specifically, if an object's account has been logged in on the terminal, the process of identity authentication is actually a process of determining whether the object using the account is consistent with the object owning the account. Therefore, the object identity information of the object owning the account can be obtained and compared with the object identity information of the object using the account. Step 205 may include:
[0151] Obtaining object identity information of a target similar object according to a similarity comparison result of the target part biometric feature, wherein the similarity between the target part biometric feature of the target similar object and the target part biometric feature of the object to be authenticated is greater than a preset biometric feature similarity threshold;
[0152] If the account of the object to be authenticated is currently logged in on the terminal, the object identity information corresponding to the currently logged in account is compared with the object identity information of the target similar object;
[0153] When the object identity information corresponding to the current login account is consistent with the object identity information of the target similar object, it is determined that the identity authentication of the object to be authenticated is successful.
[0154] The object identity information may include, but is not limited to, object ID, object nickname and / or password of the object account, etc., which can be set by the technicians according to the actual application situation.
[0155] In some optional examples, no account of any object may be logged in on the terminal. In this case, the process of identity authentication is actually to obtain the object identity information of the object (the object to be authenticated) currently being authenticated on the terminal. Therefore, step 205 may also include:
[0156] Obtaining object identity information of a target similar object according to a similarity comparison result of the target part biometric feature, wherein the similarity between the target part biometric feature of the target similar object and the target part biometric feature of the object to be authenticated is greater than a preset biometric feature similarity threshold;
[0157] The object identity information of the object to be authenticated is determined from the object identity information of the target similar object.
[0158] It is understandable that the number of target similar objects may be one or more. For example, the similarity between the target part biometric features of multiple similar objects and the target part biometric features of the object to be authenticated may be greater than a preset biometric similarity threshold.
[0159] Alternatively, the technician or the administrator of the identity authentication system can set a lower biometric similarity threshold, and select a similar object corresponding to the target part biometric feature with the highest similarity to the target part biometric feature of the object to be authenticated from at least one target part biometric feature greater than the biometric similarity threshold as the target similar object.
[0160] In one example of some optional examples, when the number of target similar objects is at least two, the authentication behavior information of the target similar objects can be obtained, and based on the object's historical authentication time, post-authentication consumption amount and / or historical authentication location in the authentication behavior, as well as the authentication behavior information of the object to be authenticated, such as the authentication time, post-authentication consumption amount, current authentication location, etc., the target similar object information with the highest similarity to the authentication behavior information of the object to be authenticated is obtained, and based on the target similar object information, the object identity information of the target similar object is obtained as the identity information of the object to be authenticated for identity authentication.
[0161] It can be understood that the identity authentication method provided by the embodiment of the present invention can be applied in many different business scenarios, and the facial feature library (similar object facial feature library, sample facial feature library and / or global facial feature library) in each business scenario can be stored in the same server or the same group of server clusters or cloud servers.
[0162] In other words, the server stores not only one facial feature library, but also multiple facial feature libraries according to specific business scenarios. For example, a shopping mall or a school should have a unique facial feature library, which contains all the objects registered in the scene. That is, the business scenario is the basic unit for object deduplication. An object may belong to multiple facial feature libraries, but each object in a facial feature library is unique.
[0163] Therefore, the similar object facial feature library corresponds to the authentication service. Before the step of "obtaining the target facial features of the object to be authenticated", the embodiment of the present invention may further include:
[0164] Obtaining an identity authentication request, the identity authentication request includes authentication service information, and the authentication service information indicates the authentication service;
[0165] Based on the authentication service information, a facial feature library of similar objects corresponding to the authentication service is determined.
[0166] Among them, the authentication business information can be a business number or a business name, etc. Different business scenarios can be distinguished by the authentication business information. The form and content of the specific authentication business information can be set by the technical personnel themselves, and the present invention does not limit this.
[0167] It should be noted that the similar object facial feature library, sample facial feature library and / or global facial feature library corresponding to each business scenario should also correspond to each other. After determining the similar object facial feature library, the sample facial feature library and / or global facial feature library corresponding to the business scenario can also be determined.
[0168] In an embodiment of the present invention, when performing identity authentication, the target facial features corresponding to the object to be authenticated should have been included in the similar object facial feature library, the sample facial feature library and / or the global facial feature library. If the target facial features corresponding to the object to be authenticated are not in the similar object facial feature library, the sample facial feature library and / or the global facial feature library, identity authentication of the object to be authenticated cannot be achieved.
[0169] Therefore, before the object to be authenticated uses the identity authentication method for the first time, the facial features corresponding to the object to be authenticated should be saved in the similar object facial feature library, the sample facial feature library and / or the global facial feature library. That is, the identity authentication method provided by the embodiment of the present invention may also include:
[0170] Obtaining the registered facial features of the registered object, and comparing the registered facial features with the global facial features in the global facial feature library for similarity;
[0171] If there is at least one global facial feature and the registered facial feature whose similarity is greater than a first facial similarity threshold, the at least one global facial feature and the registered facial feature are added to a similar object facial feature library.
[0172] The registration object is the object that uses the identity authentication device for the first time, and the registration facial features are the facial features of the registration object obtained during the registration process. It is understandable that the similar object facial feature library, the sample facial feature library and / or the global facial feature library do not contain the facial features corresponding to the registration object before obtaining the registration facial features of the registration object.
[0173] In some optional examples, the registration object can be an object that triggers object registration in the identity authentication system. For example, an identity authentication page can be displayed on the terminal, and the identity authentication page includes an identity authentication control that can trigger the identity authentication process and a registration control that can trigger the object registration process. In response to the triggering operation of the registration control, the user of the identity authentication system is determined as the registration object, and the step of obtaining the registration facial features of the registration object is executed.
[0174] In other optional examples, if facial features whose similarity with the target facial features of the object to be authenticated is greater than a preset similarity threshold are retrieved in the similar object facial feature library, the sample facial feature library and the global facial feature library, the object to be authenticated can be prompted to register. If the object to be authenticated chooses to register, the object to be authenticated is determined as a registered object, and the step of obtaining the registered facial features of the registered object is executed.
[0175] In order to ensure the information security of users, in the identity authentication method disclosed in the embodiment of the present invention, data such as facial features of similar objects, sample facial features, global facial features and / or biometric features of target parts can be stored on the blockchain.
[0176] As can be seen from the above, by adopting the scheme of the embodiment of the present invention, after obtaining the target facial features of the object to be authenticated, the facial feature library of similar objects can be preferentially retrieved. Since the facial feature library of similar objects is much smaller than the facial feature library (global facial feature library) that stores the facial features of all objects, it is possible to quickly determine whether the object to be authenticated is a similar object, thereby accelerating the identity recognition process of highly similar users belonging to similar objects, and entering the similarity comparison stage of the biometric features of the target part as soon as possible. Therefore, on the basis of ensuring the accuracy of identity authentication, the efficiency of identity authentication of the object to be authenticated is improved.
[0177] The method described in the above embodiments is further described in detail below with examples.
[0178] In this embodiment, the Figure 1 The system is described below, wherein the similar object facial feature library, the sample facial feature library and the global facial feature library are stored in the server 20.
[0179] like Figure 4 As shown, the identity authentication method of this embodiment may have the following specific process:
[0180] 401. The terminal obtains the target facial features of the object to be authenticated, and sends the target facial features to the server.
[0181] The terminal may capture an image or video stream containing the face of the object to be authenticated through a camera on the terminal or an image capture device such as an external camera or video camera.
[0182] In some embodiments, the collected image may be pre-processed at the terminal, specifically by: Figure 5 The image preprocessing module shown in is responsible for calling various algorithm SDK services to process pictures containing faces, such as liveness detection to determine whether the user is a real person, detection of whether the user is wearing a mask, and checking whether the picture quality is qualified. After passing the legality check, the features of the facial image are extracted to obtain the corresponding target facial features.
[0183] In other embodiments, the image preprocessing module may be arranged in the server, and the terminal may send the image containing the face of the object to be authenticated to the server, and the server may perform image processing.
[0184] 402. The server receives the target facial features of the object to be authenticated and reads a facial feature library of similar objects.
[0185] The similar object facial feature library stores at least one facial feature group, each facial feature group includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold.
[0186] 403. The server compares the target facial features with the facial features of similar objects in the facial feature database of similar objects in terms of facial similarity.
[0187] In the process of facial similarity comparison, the server can usually extract key facial features from one or a group of facial images sent by the terminal through a deep model and represent them as a high-dimensional vector. The server can use the similarity comparison model to compare the high-dimensional vector (target facial features) with multiple known feature vectors in the facial feature library (similar object facial feature library, sample facial feature library and / or global facial feature library) one by one to find the most similar one or more feature vectors. A common calculation method is to use the cosine distance to represent the distance between two vectors, which can also be understood as feature similarity.
[0188] In some embodiments, the facial similarity between the target facial features and the facial features of similar objects in the similar object facial feature library can be obtained by calculating the Euclidean distance, Pearson correlation coefficient, cosine similarity, etc. between the target feature vector and one or more known feature vectors.
[0189] 404. If the similarity between the target facial feature and the facial feature of at least one similar object is greater than the first facial similarity threshold, the server executes step 405; if the similarity between the target facial feature and the facial features of each similar object is not greater than the first facial similarity threshold, the server does not execute step 405 and executes step 407.
[0190] It is understandable that the similar object facial feature library stores objects with similar faces, such as twins or other people who look similar. Their characteristic is that when performing facial retrieval, multiple objects with high feature similarity are often found. Therefore, such objects are not suitable for identity authentication through facial recognition alone, which can easily lead to misidentification.
[0191] Therefore, for objects that hit the similar object facial feature library (i.e., one or more similarities exceeding the first facial similarity threshold are retrieved in the similar object facial feature library), the identity authentication system provided by the embodiment of the present invention can directly enter the palmprint recognition stage for reconfirmation.
[0192] 405. The server obtains the palm print features of the object to be authenticated, and compares the palm print features of the object to be authenticated with the palm print features of each similar object in the similar object facial feature database for similarity.
[0193] In actual application, auxiliary identification methods such as fingerprint recognition and iris recognition have disadvantages such as requiring contact and complex feature acquisition process. Non-contact palmprint recognition is easier for users to accept and does not require high collection equipment.
[0194] The object to be authenticated can extend his palm according to the prompt information displayed on the terminal, so that the image acquisition device can capture a photo of the palm. The captured photo of the palm can also be passed through Figure 5 After the palmprint features are obtained through preprocessing by the image preprocessing module in the server, Figure 5 The feature 1:1 comparison module shown is used to compare the palm print features of each similar object in the facial feature library of similar objects in the face recognition process, and obtain the similarity between the palm print features.
[0195] Among them, the feature 1:1 comparison module is responsible for comparing the two faces or palm prints in the input module and calculating the similarity.
[0196] 406 . The server determines a target similar object from the similar objects, obtains object identity information of the target similar object, and executes step 412 .
[0197] Specifically, the server can find the similar object corresponding to the most similar palm print feature among the similar objects as the target similar object. Since everyone's palm print lines are different, even for twins, their palm prints are only similar, but not exactly the same, so it can be used to make up for the possible face misrecognition problem of highly similar similar objects faced by face recognition.
[0198] 407. The server reads the sample facial feature library and compares the target facial features with the sample facial features in the sample facial feature library for facial similarity.
[0199] Specifically, the server can Figure 5 The feature 1:1 comparison module shown compares the target facial feature with each sample facial feature in the sample facial feature library, and obtains the similarity between each sample facial feature and the target facial feature.
[0200] Among them, Figure 6 As shown, there may be two levels of access queues in the sample facial feature library, including a first access queue and a second access queue.
[0201] When the facial features of an object are accessed for the first time, the facial features and the object identifier of the object can be added to the first access queue, where the first access queue adopts a first-in-first-out strategy. When the queue reaches the capacity limit (the number of sample facial features that can be stored), the sample facial features and sample object identifier of the sample object accessed the longest time ago will be eliminated from the queue.
[0202] If there is an object identifier identical to the target object identifier in the first access queue, the target object identifier and the facial features of the object to be authenticated are added to the second access queue in correspondence, and the target object identifier and the corresponding target facial features are deleted from the first access queue;
[0203] If there is an object identifier identical to the target object identifier in the second access queue, and if the second access queue reaches its capacity limit, delete the object identifier of the sample object at the end of the second access queue and the corresponding sample facial features, and move the target object identifier to the head of the second access queue.
[0204] The first access queue may include a first object identification queue and a first facial feature queue, and the first object identification queue may correspond to the first facial feature queue through object identification or storage order.
[0205] In some optional examples, the first object identification queue may not be located in the sample facial feature library, but may establish distributed storage with the first facial feature queue, be stored in different libraries, and so on.
[0206] The second access queue may have similar settings, which will not be described in detail in the present invention.
[0207] 408. If the similarity between the target facial feature and a preset number of sample facial features is greater than a preset third facial similarity threshold, the server executes step 409. If the similarity between the target facial feature and each sample facial feature is not greater than the preset third facial similarity threshold, the server executes step 410.
[0208] When the preset number is one, the sample object corresponding to the sample facial feature with the highest similarity to the target facial feature can be directly used as the target sample object.
[0209] At this time, the similarity between the sample facial features of the target sample object and the target facial features may also be no greater than the third facial similarity threshold, and so on.
[0210] 409 . The server determines a target sample object from the sample objects, obtains the object identity information of the target sample object, and executes step 412 .
[0211] When the preset number is at least two, step 409 may include:
[0212] Obtaining palm print features of the object to be authenticated, and comparing the palm print features of the object to be authenticated with palm print features of a preset number of sample objects in the sample facial feature library for similarity;
[0213] Based on the similarity comparison results of the palmprint features and the sample object information corresponding to a preset number of sample facial features, the object identity information of the target sample object is obtained.
[0214] The similarity between the palmprint features of the target sample object and the palmprint features of the object to be authenticated is the highest among the palmprint features of a preset number of sample objects.
[0215] 410. The server reads the global facial feature library and compares the target facial features with the global facial features in the global facial feature library for facial similarity.
[0216] Since the global facial feature database may contain a large amount of data, accurate similarity calculations one by one require high computing resources. Therefore, we can first use Figure 5 The feature 1:N retrieval module shown in the figure calculates the similarity of an input facial feature (target facial feature) with all global facial features in the global facial feature library, and finds the one or top k features with the highest similarity. At this time, the calculated similarity may have a low accuracy. For example, the similarity can be calculated only based on data such as facial contours and facial features contours, and so on.
[0217] Then, the input target facial features are compared one by one with the 1:1 feature comparison module and the 1 or k global facial features obtained by the 1:N feature retrieval module to calculate the similarity.
[0218] In some optional embodiments, the feature 1:N retrieval module can batch calculate the target facial features and the global facial features in the global facial feature library by means of matrix calculation, etc., and can improve the speed of facial similarity comparison on the basis of ensuring the accuracy of similarity. In this case, the facial similarity comparison of the target facial features and the global facial features in the global facial feature library can be performed only by the feature 1:N retrieval module.
[0219] 411. The server determines a target global object from the global objects and obtains object identity information of the target global object.
[0220] Specifically, the global object corresponding to the global facial feature having the highest similarity to the target facial feature may be directly used as the target global object.
[0221] The server can be Figure 5 The storage layer shown in obtains the object identity information corresponding to the target global object.
[0222] 412. The server performs identity authentication on the object to be authenticated according to the object identity information, and returns the identity authentication result to the terminal.
[0223] like Figure 5 The identity authentication system shown can be divided into three layers: access layer, logic layer, and storage layer.
[0224] Among them, the access layer can include two access modules: object management and library management. The object management module is responsible for providing interfaces such as object-related identity information registration, deletion, and update, and the library management module is responsible for providing interfaces such as object identity recognition, N library creation, deletion, and modification.
[0225] Among them, the logic layer can be responsible for implementing specific business logic and serial algorithm processes, mainly including image preprocessing, object cache module, feature 1:1 comparison, feature 1:N retrieval and multiple algorithm SDK microservice modules, such as facial feature extraction service, palmprint feature extraction service, facial liveness detection service, palmprint liveness detection service, etc.
[0226] Among them, the storage layer is mainly responsible for data storage, including object identity information library, facial feature N library, palm print feature N library, etc. The object identity information library stores detailed information of all registered objects, and the facial or palm print feature N library stores feature data of registered objects, and each feature data is associated with the object identity information one by one.
[0227] As can be seen from the above, after acquiring the target facial features of the object to be authenticated, the embodiment of the present invention preferentially searches the facial feature library of similar objects. Since the facial feature library of similar objects is much smaller than the facial feature library (global facial feature library) that stores the facial features of all objects, it can quickly determine whether the object to be authenticated is a similar object, thereby accelerating the identity recognition process of highly similar users belonging to similar objects, and entering the similarity comparison stage of the biometric features of the target part as soon as possible. Therefore, on the basis of ensuring the accuracy of identity authentication, the efficiency of identity authentication of the object to be authenticated is improved.
[0228] In order to better implement the above method, accordingly, an embodiment of the present invention further provides an identity authentication device.
[0229] refer to Figure 7 , the device may include:
[0230] The facial feature acquisition unit 701 may be used to acquire the target facial features of the object to be authenticated and read the similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each facial feature group may include similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold;
[0231] The feature comparison unit 702 may be used to compare the target facial features with the facial features of similar objects in the similar object facial feature library for facial similarity;
[0232] The biometric feature acquisition unit 703 may be configured to acquire the biometric feature of the target part of the object to be authenticated if the similarity between the target facial feature and the facial feature of at least one similar object is greater than a first facial similarity threshold;
[0233] The biometric feature comparison unit 704 may be used to compare the biometric features of the target part of the object to be authenticated with the biometric features of the target part of each similar object in the similar object facial feature library for similarity;
[0234] The identity authentication unit 705 may be used to perform identity authentication on the object to be authenticated based on the similarity comparison result of the biometric features of the target part.
[0235] Optional, such as Figure 8 As shown, the identity authentication device may also include:
[0236] The global feature comparison unit 706 may be configured to compare the target facial feature with the global facial features in the global facial feature library for facial similarity if the similarity between the target facial feature and the facial features of each similar object is not greater than a first facial similarity threshold, wherein the global facial feature library may include global facial features of multiple global objects;
[0237] The second identity authentication unit 707 may be used to perform identity authentication on the object to be authenticated based on the result of the facial similarity comparison.
[0238] Optionally, the second identity authentication unit 707 may be configured to obtain a target part biometric feature of the object to be authenticated if the similarity between the target facial feature and at least two global facial features is greater than a second facial similarity threshold;
[0239] Based on the target part biometric features of the object to be authenticated and the target part biometric features of each global object in the global facial feature library, a similarity comparison is performed on the target part biometric features;
[0240] Based on the similarity comparison result of the biometric features of the target part, the identity of the object to be authenticated is authenticated.
[0241] Optionally, the second identity authentication unit 707 may be used to determine the target part biometric feature of the target global object from the target part biometric features of each global object in the global facial feature library, and the similarity between the global facial feature of the target global object and the target facial feature is greater than a second facial similarity threshold;
[0242] The target part biometric features of the object to be authenticated are compared with the target part biometric features of each target global object for similarity.
[0243] Optionally, the identity authentication device may further include:
[0244] The global object identity authentication unit 708 can be used to authenticate the object to be authenticated based on the global object information corresponding to the target number of global facial features if the similarity between the target facial feature and the target number of global facial features in the global facial feature library is greater than a second facial similarity threshold.
[0245] Optionally, before the global feature comparison unit 706, the following may also be included:
[0246] The sample feature comparison unit 709 may be used to obtain a sample facial feature library, wherein the sample objects in the sample facial feature library have undergone identity authentication at least once;
[0247] Compare the target facial features with the sample facial features in the sample facial feature library for facial similarity;
[0248] If the similarity between the target facial feature and each sample facial feature is not greater than a preset third facial similarity threshold, a step of comparing the facial similarity between the target facial feature and the global facial features in the global facial feature library is performed.
[0249] Optionally, the sample feature comparison unit 709 may also be used to: if the similarity between the target facial feature and a preset number of sample facial features is greater than a preset third facial similarity threshold, perform identity authentication on the object to be authenticated based on sample object information corresponding to the preset number of sample facial features.
[0250] Optionally, the sample facial feature library may include a first access queue, and the number of sample facial features that can be stored in the first access queue is certain. The second identity authentication unit 707 may also include a first feature storage unit 710, which can be used to add the target object identifier of the object to be authenticated and the target facial feature of the object to be authenticated to the first access queue, and delete the object identifier of the sample object with the longest storage time in the first access queue and the corresponding sample facial feature.
[0251] Optionally, a second feature storage unit 711 may be included before the first feature storage unit 710, which may be used to add the target object identifier and the facial features of the object to be authenticated to the second access queue if there is an object identifier identical to the target object identifier in the first access queue, and delete the target object identifier and the corresponding target facial features from the first access queue;
[0252] If there is an object identifier identical to the target object identifier in the second access queue, the target object identifier is moved to the head of the second access queue.
[0253] Optionally, the facial feature library of similar objects corresponds to the authentication service, and the facial feature acquisition unit 701 may further include a feature library determination unit 712, which may be used to obtain an identity authentication request, and the identity authentication request may include authentication service information, and the authentication service information indicates the authentication service;
[0254] Based on the authentication service information, a facial feature library of similar objects corresponding to the authentication service is determined.
[0255] Optionally, the identity authentication device provided by the embodiment of the present invention may further include a registration unit 713, which may be used to obtain the registered facial features of the registered object, and compare the similarity between the registered facial features and each global facial feature in the global facial feature library;
[0256] If there is at least one global facial feature and the registered facial feature whose similarity is greater than a first facial similarity threshold, the at least one global facial feature and the registered facial feature are added to a similar object facial feature library.
[0257] Optionally, the facial feature acquisition unit 701 may further include an identification acquisition unit 714, which may be used to acquire a similar identification of the object to be authenticated, where the similar identification indicates whether the object to be authenticated is a similar object;
[0258] When the similarity identifier indicates that the object to be authenticated is a similar object, the step of acquiring the target facial features of the object to be authenticated is not performed, and the step of acquiring the target part biometric features of the object to be authenticated is performed.
[0259] As can be seen from the above, the embodiment of the present invention uses the identity authentication device to preferentially search the similar object facial feature library after obtaining the target facial features of the object to be authenticated. Since the similar object facial feature library is much smaller than the facial feature library (global facial feature library) that stores the facial features of all objects, it can quickly determine whether the object to be authenticated is a similar object, thereby accelerating the identity recognition process of highly similar users belonging to similar objects, and entering the similarity comparison stage of the target part biometric features as soon as possible. Therefore, on the basis of ensuring the accuracy of identity authentication, the efficiency of identity authentication of the object to be authenticated is improved.
[0260] In addition, an embodiment of the present invention further provides an electronic device, which may be a terminal or a server, etc. Fig. 9 As shown, it shows a schematic diagram of the structure of an electronic device involved in an embodiment of the present invention, specifically:
[0261] The electronic device may include a radio frequency (RF) circuit 901, a memory 902 including one or more computer-readable storage media, an input unit 903, a display unit 904, a sensor 905, an audio circuit 906, a wireless fidelity (WiFi) module 907, a processor 908 including one or more processing cores, and a power supply 909. Those skilled in the art will appreciate that Fig. 9 The electronic device structure shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0262] in:
[0263] The RF circuit 901 can be used for receiving and sending signals during the process of sending and receiving information or making calls. In particular, after receiving the downlink information of the base station, it is handed over to one or more processors 908 for processing; in addition, the data related to the uplink is sent to the base station. Usually, the RF circuit 901 includes but is not limited to an antenna, at least one amplifier, a tuner, one or more oscillators, a subscriber identity module (SIM) card, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, etc. In addition, the RF circuit 901 can also communicate with the network and other devices through wireless communication. Wireless communication can use any communication standard or protocol, including but not limited to the Global System of Mobile communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), etc.
[0264] The memory 902 can be used to store software programs and modules, and the processor 908 executes various functional applications and data processing by running the software programs and modules stored in the memory 902. The memory 902 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device (such as audio data, a phone book, etc.), etc. In addition, the memory 902 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 902 may also include a memory controller to provide the processor 908 and the input unit 903 with access to the memory 902.
[0265] The input unit 903 can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control. Specifically, in a specific embodiment, the input unit 903 may include a touch-sensitive surface and other input devices. The touch-sensitive surface, also known as a touch display screen or a touch pad, can collect the user's touch operations on or near it (such as the user's operation on or near the touch-sensitive surface using any suitable object or accessory such as a finger, a stylus, etc.), and drive the corresponding connection device according to a pre-set program. Optionally, the touch-sensitive surface may include a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into the touch point coordinates, and then sends it to the processor 908, and can receive and execute the command sent by the processor 908. In addition, the touch-sensitive surface can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic waves. In addition to the touch-sensitive surface, the input unit 903 may also include other input devices. Specifically, other input devices may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.
[0266] The display unit 904 can be used to display information input by the user or information provided to the user and various graphical user interfaces of the electronic device, which can be composed of graphics, text, icons, videos and any combination thereof. The display unit 904 may include a display panel. Optionally, the display panel may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. Further, the touch-sensitive surface may cover the display panel. When the touch-sensitive surface detects a touch operation on or near it, it is transmitted to the processor 908 to determine the type of touch event. Subsequently, the processor 908 provides a corresponding visual output on the display panel according to the type of touch event. Although in Figure 6 In the embodiment, the touch-sensitive surface and the display panel are used as two independent components to realize input and output functions, but in some embodiments, the touch-sensitive surface and the display panel can be integrated to realize input and output functions.
[0267] The electronic device may also include at least one sensor 905, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel according to the brightness of the ambient light, and the proximity sensor may turn off the display panel and / or backlight when the electronic device is moved to the ear. As a type of motion sensor, the gravity acceleration sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used for applications that identify the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors that can be configured in the electronic device, such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., they will not be repeated here.
[0268] The audio circuit 906, the speaker, and the microphone can provide an audio interface between the user and the electronic device. The audio circuit 906 can transmit the electrical signal converted from the received audio data to the speaker, which is converted into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 906 and converted into audio data, and then the audio data is output to the processor 908 for processing, and then sent to another electronic device through the RF circuit 901, or the audio data is output to the memory 902 for further processing. The audio circuit 906 may also include an earplug jack to provide communication between an external headset and the electronic device.
[0269] WiFi is a short-range wireless transmission technology. Electronic devices can help users send and receive emails, browse web pages, and access streaming media through WiFi module 907. It provides users with wireless broadband Internet access. Figure 6 A WiFi module 907 is shown, but it is understandable that it is not an essential component of the electronic device and can be omitted as needed without changing the essence of the invention.
[0270] The processor 908 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and / or modules stored in the memory 902, and calling data stored in the memory 902, it executes various functions of the electronic device and processes data, thereby performing overall detection of the mobile phone. Optionally, the processor 908 may include one or more processing cores; preferably, the processor 908 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 908.
[0271] The electronic device also includes a power supply 909 (such as a battery) for supplying power to each component. Preferably, the power supply can be logically connected to the processor 908 through a power management system, so that the power management system can manage charging, discharging, and power consumption management. The power supply 909 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.
[0272] Although not shown, the electronic device may also include a camera, a Bluetooth module, etc., which will not be described in detail here. Specifically in this embodiment, the processor 908 in the electronic device will load the executable files corresponding to the processes of one or more applications into the memory 902 according to the following instructions, and the processor 908 will run the application stored in the memory 902 to implement various functions, as follows:
[0273] Acquire the target facial features of the object to be authenticated and read the similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each facial feature group includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold;
[0274] Comparing the facial similarity between the target facial features and the facial features of similar objects in the facial feature library of similar objects;
[0275] If the similarity between the target facial feature and the facial feature of at least one similar object is greater than a first facial similarity threshold, obtaining a target part biometric feature of the object to be authenticated;
[0276] Compare the biometric features of the target part of the object to be authenticated with the biometric features of the target part of each similar object in the facial feature library of similar objects;
[0277] Based on the similarity comparison result of the biometric features of the target part, the identity of the object to be authenticated is authenticated.
[0278] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.
[0279] To this end, an embodiment of the present invention provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any identity authentication method provided in the embodiment of the present invention. For example, the instructions can execute the following steps:
[0280] Acquire the target facial features of the object to be authenticated and read the similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each facial feature group includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold;
[0281] Comparing the facial similarity between the target facial features and the facial features of similar objects in the facial feature library of similar objects;
[0282] If the similarity between the target facial feature and the facial feature of at least one similar object is greater than a first facial similarity threshold, obtaining a target part biometric feature of the object to be authenticated;
[0283] Compare the target part biometric features of the object to be authenticated with the target part biometric features of each similar object in the similar object facial feature library;
[0284] Based on the similarity comparison result of the biometric features of the target part, the identity of the object to be authenticated is authenticated.
[0285] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.
[0286] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.
[0287] Since the instructions stored in the computer-readable storage medium can execute the steps in any identity authentication method provided in the embodiments of the present invention, the beneficial effects that can be achieved by any identity authentication method provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.
[0288] According to one aspect of the present application, a computer program product or a computer program is also provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the methods provided in various optional implementations of the above embodiments.
[0289] The above is a detailed introduction to an identity authentication method, device, electronic device and storage medium provided in an embodiment of the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
Claims
1. An identity authentication method, characterized in that: include: Acquire the target facial features of the object to be authenticated and read a similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each of the facial feature groups includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold; Comparing the target facial features with the facial features of similar objects in the similar object facial feature library in terms of facial similarity; If the similarity between the target facial feature and at least one of the similar object facial features is greater than the first facial similarity threshold, acquiring the target part biometric feature of the object to be authenticated; Comparing the target part biometric features of the object to be authenticated with the target part biometric features of each of the similar objects in the similar object facial feature library for similarity; Based on the similarity comparison result of the biological characteristics of the target part, the identity of the object to be authenticated is authenticated; If the similarity between the target facial feature and the facial features of each of the similar objects is not greater than the first facial similarity threshold, obtaining a sample facial feature library; the sample facial feature library includes a first access queue and a second access queue; Comparing the target facial features with the sample facial features in the sample facial feature library for facial similarity, so as to perform identity authentication on the object to be authenticated; The first access queue is obtained by: adding the target object identifier of the object to be authenticated and the target facial feature of the object to be authenticated to the first access queue accordingly, and deleting the object identifier of the sample object with the longest storage time and the corresponding sample facial feature in the first access queue; The second access queue is obtained in the following manner: if there is an object identifier identical to the target object identifier in the first access queue, the target object identifier and the target facial feature of the object to be authenticated are added to the second access queue accordingly, and the target object identifier and the corresponding target facial feature are deleted from the first access queue.
2. The identity authentication method according to claim 1, characterized in that: The method further comprises: If the similarity between the target facial feature and the facial features of each of the similar objects is not greater than the first facial similarity threshold, comparing the target facial feature with the global facial features in the global facial feature library for facial similarity, wherein the global facial feature library includes global facial features of multiple global objects; Based on the result of the facial similarity comparison, identity authentication is performed on the object to be authenticated.
3. The identity authentication method according to claim 2, characterized in that: The step of performing identity authentication on the object to be authenticated based on the result of the facial similarity comparison includes: If the similarity between the target facial feature and at least two of the global facial features is greater than a second facial similarity threshold, obtaining a target part biometric feature of the object to be authenticated; Based on the target part biometric features of the object to be authenticated and the target part biometric features of each of the global objects in the global facial feature library, performing a similarity comparison of the target part biometric features; Based on the similarity comparison result of the biological features of the target part, the identity of the object to be authenticated is authenticated.
4. The method according to claim 3, characterized in that The similarity comparison of the target part biometric features based on the target part biometric features of the object to be authenticated and the target part biometric features of each global object in the global facial feature library includes: Determining a target part biometric feature of a target global object from the target part biometric features of each of the global objects in the global facial feature library, wherein the similarity between the global facial feature of the target global object and the target facial feature is greater than the second facial similarity threshold; The target part biometric feature of the object to be authenticated is compared with the target part biometric features of each target global object for similarity.
5. The identity authentication method according to claim 2, characterized in that: The step of performing identity authentication on the object to be authenticated based on the result of the facial similarity comparison includes: If the similarity between the target facial feature and a target number of the global facial features in the global facial feature library is greater than a second facial similarity threshold, identity authentication is performed on the object to be authenticated based on the global object information corresponding to the target number of global facial features.
6. The identity authentication method according to claim 2, characterized in that: Before comparing the target facial features with the global facial features in the global facial feature library for facial similarity, the method further includes: Obtaining a sample facial feature library, wherein the sample objects in the sample facial feature library have undergone identity authentication at least once; Comparing the target facial features with the sample facial features in the sample facial feature library for facial similarity; If the similarity between the target facial feature and each of the sample facial features is not greater than a preset third facial similarity threshold, the step of comparing the facial similarity between the target facial feature and the global facial features in the global facial feature library is performed.
7. The identity authentication method according to claim 6, characterized in that: The method further comprises: If the similarity between the target facial feature and a preset number of sample facial features is greater than a preset third facial similarity threshold, the identity of the object to be authenticated is performed based on the sample object information corresponding to the preset number of sample facial features.
8. The identity authentication method according to claim 6, characterized in that: The first access queue can store a certain number of sample facial features.
9. The identity authentication method according to claim 8, characterized in that: The method further comprises: If there is an object identifier identical to the target object identifier in the second access queue, the target object identifier is moved to the head of the second access queue.
10. The identity authentication method according to claim 1, characterized in that: The similar object facial feature library corresponds to the authentication service; Before acquiring the target facial features of the object to be authenticated, the method further includes: Obtaining an identity authentication request, wherein the identity authentication request includes authentication service information, and the authentication service information indicates an authentication service; Based on the authentication service information, a facial feature library of similar objects corresponding to the authentication service is determined.
11. The identity authentication method according to any one of claims 2 to 9, characterized in that: The method further comprises: Acquire the registered facial features of the registered object, and compare the similarity between the registered facial features and the global facial features in the global facial feature library; If there is at least one of the global facial features and the registered facial feature whose similarity is greater than the first facial similarity threshold, the at least one global facial feature and the registered facial feature are added to the similar object facial feature library.
12. The identity authentication method according to any one of claims 1 to 10, characterized in that: Before obtaining the target facial features of the object to be authenticated, the method further includes: Acquire a similarity identifier of the object to be authenticated, wherein the similarity identifier indicates whether the object to be authenticated is a similar object; When the similarity identifier indicates that the object to be authenticated is a similar object, the step of acquiring the target facial features of the object to be authenticated is not performed, and the step of acquiring the target part biometric features of the object to be authenticated is performed.
13. An identity authentication device, characterized in that: include: A facial feature acquisition unit, used to acquire a target facial feature of an object to be authenticated and read a similar object facial feature library, wherein the similar object facial feature library stores at least one facial feature group, each of the facial feature groups includes similar object facial features of at least two similar objects, and the facial feature similarity between the similar objects in each facial feature group is greater than a first facial similarity threshold; A feature comparison unit, used for comparing the target facial features with the facial features of similar objects in the similar object facial feature library in terms of facial similarity; a biometric feature acquisition unit, configured to acquire a target part biometric feature of the object to be authenticated if the similarity between the target facial feature and at least one facial feature of the similar object is greater than the first facial similarity threshold; A biometric feature comparison unit, used for comparing the biometric features of the target part of the object to be authenticated with the biometric features of the target part of each of the similar objects in the similar object facial feature library for similarity; The identity authentication unit is used to authenticate the identity of the object to be authenticated based on the similarity comparison result of the biological features of the target part; and is also used to obtain a sample facial feature library if the similarity between the target facial features and the facial features of each of the similar objects is not greater than the first facial similarity threshold; the sample facial feature library includes a first access queue and a second access queue; the target facial features are compared with the sample facial features in the sample facial feature library for facial similarity to authenticate the identity of the object to be authenticated; The first access queue is obtained by: adding the target object identifier of the object to be authenticated and the target facial feature of the object to be authenticated to the first access queue accordingly, and deleting the object identifier of the sample object with the longest storage time and the corresponding sample facial feature in the first access queue; The second access queue is obtained in the following manner: if there is an object identifier identical to the target object identifier in the first access queue, the target object identifier and the target facial feature of the object to be authenticated are added to the second access queue accordingly, and the target object identifier and the corresponding target facial feature are deleted from the first access queue.
14. An electronic device, characterized in that: It comprises a memory and a processor; the memory stores an application program, and the processor is used to run the application program in the memory to execute the steps in the identity authentication method described in any one of claims 1 to 12.
15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the identity authentication method described in any one of claims 1 to 12.
Citation Information
Patent Citations
Identity information retrieval method and device, service system and electronic equipment
CN110991390A
Identity recognition method and device
CN111291645A
Method and device for carrying out user verification based on feature information
CN111506889A
Identity authentication method and device, electronic equipment and readable storage medium
CN111709004A
Facial recognition method, apparatus and system, device, and storage medium
WO2017181769A1
