Identity identification method, device, computer equipment and storage medium

Abstract

The present application relates to an identity recognition method, device, computer equipment, storage medium and computer program product. The method comprises: in response to an identity recognition triggering event, for a target part of a target user, collecting a combination of biometric features when the target part is in a partially hidden deformed form; the biometric feature combination is a combination of at least two types of biometric features at the target part; matching the collected biometric feature combination with a pre-stored registered biometric feature pattern to obtain a biometric feature matching result; the registered biometric feature pattern is a combination of biometric features obtained by identity registration of the target part of the registered user when the target part of the registered user is in a partially hidden deformed form; and determining the identity recognition result for the target user based on the biometric feature matching result. The use of this method can improve the security of biometric features in the identity recognition process.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an identity recognition method, apparatus, computer equipment, storage medium and computer program product. Background Art

[0002] With the development of computer technology, increasingly mature identity recognition technology has been widely used in business cooperation, consumer payment, social media, security and other fields. Among them, using people's inherent biological characteristics, such as hand shape, fingerprint, face shape, retina, auricle and other biological characteristics for identity recognition has become the development trend of identity recognition technology.

[0003] However, in the current technology of identity recognition based on biometrics, when a user uses biometrics such as hand shape, face, fingerprint, palm print, etc. for identity recognition, if the biometrics are stolen, such as the hand shape, face, fingerprint or palm print is photographed and stolen by others, there is a security risk that others may forge the biometrics, and the security of biometrics is low during the identity recognition process. Summary of the invention

[0004] Based on this, it is necessary to provide an identity recognition method, device, computer equipment, storage medium and computer program product that can improve the security of biometrics in the identity recognition process in response to the above technical problems.

[0005] An identity recognition method, the method comprising:

[0006] In response to an identity recognition trigger event, for a target part of a target user, a combination of biometric features is collected when the target part is in a partially hidden deformed state; the combination of biometric features is a combination of at least two types of biometric features at the target part;

[0007] The collected biometric combination is feature matched with the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form;

[0008] Based on the biometric matching results, the identity recognition result for the target user is determined.

[0009] An identity recognition device, comprising:

[0010] A biometric feature collection module, for responding to an identity recognition trigger event, for a target part of a target user, collecting a biometric feature combination when the target part is in a partially hidden deformed state; the biometric feature combination is a combination of at least two types of biometric features at the target part;

[0011] The biometric matching module is used to match the collected biometric combination with the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form;

[0012] The identity recognition processing module is used to determine the identity recognition result for the target user based on the biometric matching result.

[0013] In one embodiment, the target part is a palm; the biometric feature combination includes at least a palm print feature and a palm vein feature at the palm; the biometric feature matching module includes: a registration mode determination module, used to determine a pre-stored registration biometric feature mode; the registration biometric feature mode includes at least a palm print registration feature and a palm vein registration feature obtained by identity registration of the palm of a registered user; a palm print matching module, used to perform palm print feature matching on the palm print registration feature of the collected palm print feature in the biometric feature combination to obtain a palm print feature matching result; a palm vein matching module, used to perform palm vein feature matching on the palm vein registration feature of the collected palm vein feature in the biometric feature combination to obtain a palm vein feature matching result; and a matching result obtaining module, used to obtain a biometric feature matching result based on the palm print feature matching result and the palm vein feature matching result.

[0014] In one of the embodiments, the matching result obtaining module includes: a matching weight module, used to determine the palm print matching weight and the palm vein matching weight; a weighted fusion module, used to perform weighted fusion on the palm print feature matching result and the palm vein feature matching result based on the palm print matching weight and the palm vein matching weight to obtain a biometric matching result.

[0015] In one of the embodiments, the target part is a palm; the biometric feature acquisition module includes: an identification trigger response module, which is used to enter a palm feature acquisition state in response to an identity recognition trigger event; a feature map acquisition module, which is used to collect a biometric map of the target user's palm in the palm feature acquisition state; and a feature extraction module, which is used to extract a biometric feature combination of the target user's palm from the biometric map when it is determined that the target user's palm is in a partially hidden palm shape.

[0016] In one of the embodiments, the feature extraction module includes: a morphology recognition module, used to perform palm morphology recognition on the palm of the target user based on the biometric map to obtain a palm morphology recognition result; a palm feature extraction module, used to extract at least two types of palm biometric features of the palm of the target user from the biometric map when the palm morphology recognition result indicates that the palm of the target user is in a partially hidden palm shape; and a palm feature combination acquisition module, used to obtain a biometric feature combination of the palm of the target user based on the at least two types of palm biometric features.

[0017] In one of the embodiments, the device also includes: a gesture determination module, used to determine the target gesture corresponding to the palm of the target user according to the biometric map; a gesture recognition module, used to perform gesture recognition on the target gesture based on each registered gesture in a local registered gesture library to obtain a gesture recognition result; the biometric matching module is also used to send the collected biometric combination to the server when the gesture recognition result is recognition passed, so that the server can perform feature matching on the collected biometric combination with a pre-stored registered biometric pattern; and obtain the biometric matching result returned by the server.

[0018] In one of the embodiments, the device also includes: a registration event response module, which is used to enter a biometric collection state in response to a triggering event of registration identity recognition; a registration feature acquisition module, which is used to collect the biometric combination to be registered when the target part of the registered user is in a partially hidden deformed form in the biometric collection state; and a registration processing module, which is used to perform identity registration on the target part of the registered user based on the biometric combination to be registered when the biometric combination to be registered meets the identity recognition registration conditions, and obtain the registered biometric pattern.

[0019] In one of the embodiments, the device also includes: a replacement event response module, which is used to respond to a triggering event of replacing a registered biometric pattern, and collect the biometric combination to be replaced when the target part of the registered user is in a partially hidden deformed form; a replacement feature acquisition module, which is used to perform feature validity analysis on various types of biometrics in the biometric combination to be replaced to obtain a feature analysis result; and a replacement processing module, which is used to replace the biometric combination in the registered biometric pattern with the biometric combination to be replaced when the feature analysis result indicates that the biometric combination to be replaced is valid.

[0020] In one of the embodiments, the device also includes: a resource transfer parameter determination module, used to determine the resource transfer parameters in response to a resource transfer trigger event; a resource account determination module, used to determine the target resource account associated with the target user based on the identity recognition result; and a resource transfer processing module, used to transfer resources to the target resource account based on the resource transfer parameters.

[0021] In one embodiment, the biometric matching module includes: a feature pattern determination module, used to determine a pre-stored registered biometric pattern; a feature pattern matching module, used to perform feature matching on each type of biometric in the collected biometric combination with the registered biometric of the corresponding type of each type of biometric in the registered biometric pattern to obtain feature matching results corresponding to each type of biometric; and a feature matching result acquisition module, used to obtain a biometric matching result based on the feature matching results corresponding to each type of biometric.

[0022] In one of the embodiments, the biometric matching module includes: a scene determination module, used to determine the trigger scene corresponding to the identity recognition trigger event; a scene feature pattern determination module, used to determine the target registered biometric pattern associated with the trigger scene from the pre-stored registered biometric patterns; a scene feature pattern matching module, used to combine the collected biometric features and perform feature matching with the target registered biometric pattern to obtain a biometric matching result.

[0023] In one of the embodiments, the registered biometric pattern includes at least two biometric combinations obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in different deformed forms of local concealment; the biometric matching module is further used to perform feature matching on the collected biometric combination with various biometric combinations in the pre-stored registered biometric pattern to obtain biometric matching results corresponding to each biometric combination in the registered biometric pattern; the identity recognition processing module is further used to determine the identity recognition result for the target user based on the successfully matched biometric matching results among the biometric matching results corresponding to the various biometric combinations.

[0024] In one of the embodiments, the device further includes: a feature image acquisition module, used to acquire a feature image of the target part of the target user in response to an identity recognition trigger event; a local hidden analysis module, used to perform a local hidden analysis on the target part based on the feature image to obtain a local hidden analysis result; and an identification failure determination module, used to determine that the identity recognition result of the target user is an identity recognition failure when the local hidden analysis result indicates that the target part is not in a locally hidden deformed form.

[0025] In one of the embodiments, the device also includes: a depth image acquisition module, which is used to collect a depth image corresponding to the target part of the target user in response to an identity recognition trigger event; a liveness detection module, which is used to perform liveness detection on the target user based on the depth image to obtain a liveness detection result; and a detection result processing module, which is used to determine that the identity recognition result of the target user is an identity recognition failure when the liveness detection result is a detection failure.

[0026] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0027] In response to an identity recognition trigger event, for a target part of a target user, a combination of biometric features is collected when the target part is in a partially hidden deformed state; the combination of biometric features is a combination of at least two types of biometric features at the target part;

[0028] The collected biometric combination is feature matched with the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form;

[0029] Based on the biometric matching results, the identity recognition result for the target user is determined.

[0030] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0031] In response to an identity recognition trigger event, for a target part of a target user, a combination of biometric features is collected when the target part is in a partially hidden deformed state; the combination of biometric features is a combination of at least two types of biometric features at the target part;

[0032] The collected biometric combination is feature matched with the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form;

[0033] Based on the biometric matching results, the identity recognition result for the target user is determined.

[0034] A computer program product. The computer program product comprises a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0035] In response to an identity recognition trigger event, for a target part of a target user, a combination of biometric features is collected when the target part is in a partially hidden deformed state; the combination of biometric features is a combination of at least two types of biometric features at the target part;

[0036] The collected biometric combination is feature matched with the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form;

[0037] Based on the biometric matching results, the identity recognition result for the target user is determined.

[0038] In the above-mentioned identity recognition method, apparatus, computer equipment, storage medium and computer program product, during the identity recognition process, the biometric feature combination collected for feature matching is a combination of at least two types of biometric features at the target part of the target user when the target part of the target user is in a partially hidden deformed form, and the pre-stored registered biometric feature pattern is a biometric feature combination obtained by identity registration of the target part of the registered user when the target part of the registered user is in a partially hidden deformed form. Identity recognition is performed based on the collected biometric feature combination and the pre-stored registered biometric feature pattern, which can prevent the global biometric features of the user's target part from being stolen, thereby improving the security of the user's biometric features during the identity recognition process. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A diagram of an application environment of an identity recognition method in an embodiment;

[0040] Figure 2 is a flow chart of an identity recognition method in one embodiment;

[0041] Figure 3 A schematic diagram of a flow chart of feature matching processing in one embodiment;

[0042] Figure 4 A schematic diagram of palm prints in one embodiment;

[0043] Figure 5 is a schematic diagram of palm veins in one embodiment;

[0044] Figure 6 is a schematic diagram of a vertical palm brushing device in one embodiment;

[0045] Figure 7 is a schematic diagram of a horizontal palm brushing device in one embodiment;

[0046] Figure 8 A schematic diagram of a hand shape in which a part of the hand is hidden by the middle finger, the ring finger and the little finger in one embodiment;

[0047] Fig. 9 A schematic diagram of a hand shape in which a thumb is used to partially hide a part of the hand in one embodiment;

[0048] Fig.10 A schematic diagram of a hand shape in which a part is hidden by an index finger in one embodiment;

[0049] Fig.11 A schematic diagram of a hand shape in which a portion of the finger is partially hidden by using the index finger, the middle finger, the ring finger and the little finger in one embodiment;

[0050] Fig.12 A schematic diagram of a hand shape in which a part of the hand is partially hidden by indenting the palm in an embodiment;

[0051] Fig.13 A schematic diagram of the hand shape of an orchid finger in one embodiment;

[0052] Fig.14 is a schematic diagram of a process of registering a hand shape mode in one embodiment;

[0053] Fig.15 A schematic diagram of a palm brushing process in one embodiment;

[0054] Fig.16 It is a schematic diagram of a process of updating a hand shape mode in one embodiment;

[0055] Fig.17 is a schematic diagram of a field of view angle in one embodiment;

[0056] Fig.18 is a schematic diagram of depth of field in one embodiment;

[0057] Fig.19 This is a schematic diagram of the traditional palm-swipe payment scenario;

[0058] Fig. 20 A schematic diagram of a scenario of palm payment by partially hiding a hand shape in one embodiment;

[0059] Fig.21 A schematic diagram of a scenario of palm payment by partially hiding a hand shape in another embodiment;

[0060] Fig. 22 is a structural block diagram of an identity recognition device in an embodiment;

[0061] Fig.23 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0063] The identity recognition method provided by this application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. In response to the identity recognition trigger event, the terminal 102 collects the biometric combination of the target part of the target user when the target part is in a partially hidden deformed form. The biometric combination is a combination of at least two types of biometrics at the target part. The terminal 102 sends the collected biometric combination to the server 104, so that the server 104 performs feature matching on the collected biometric combination with the pre-stored registered biometric pattern to obtain a biometric matching result, and the server 104 returns the biometric matching result to the terminal 102. The registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form. The terminal 102 determines the identity recognition result for the target user based on the biometric matching result returned by the server 104, thereby realizing the identity recognition of the target user. In addition, in specific applications, the processing of identity recognition can also be implemented by the terminal 102 alone, that is, the terminal 102 collects the biometric combination of the target user when the target part is in a partially hidden deformed form, and matches the biometric combination with the pre-stored registered biometric pattern, and determines the identity recognition result for the target user based on the biometric matching result.

[0064] Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The terminal 102 can be configured with sensor devices for collecting various types of biometric features of the target parts of the user to achieve the collection of different types of biometric features of the target parts. The server 104 can be implemented as an independent server or a server cluster composed of multiple servers. As disclosed in the identity recognition method of the present application, the multiple servers involved can be composed of a blockchain, and the server is a node on the blockchain.

[0065] In one embodiment, Figure 2 As shown, an identity recognition method is provided, which is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0066] Step 202, in response to an identity recognition triggering event, for a target part of a target user, a biometric feature combination is collected when the target part is in a partially hidden deformed state; the biometric feature combination is a combination of at least two types of biometric features at the target part.

[0067] Among them, identity recognition is the verification process of identifying whether the real identity of the user is consistent with the identity he claims. With the development of identity recognition technology, identity recognition methods based on biometrics have been widely used. Identity recognition trigger events refer to events that trigger identity recognition, which may include but are not limited to operations and instructions that trigger identity recognition. For example, in the access control system scenario, when a user needs to pass through the access control, an identity recognition event is triggered; for example, when a user makes a payment at a payment terminal, an identity recognition event is triggered. In addition, identity recognition can also be applied to anti-addiction system scenarios. For example, in the online game anti-addiction system, it is necessary to limit the online game time of minors. When anti-addiction is triggered, if the cumulative online game time of the game user reaches the preset time threshold, the game user needs to be identified. At this time, an identity recognition event is triggered to determine whether the game user is an adult or the game account owner, thereby limiting the online game time of minors. In a specific implementation, an identity recognition trigger event is an event that triggers identity recognition through biometrics. Biometrics are biometrics of body parts that can be measured by users, such as various types of biometrics such as hand shape, fingerprint, face shape, iris, retina, palm, etc. When performing identity recognition processing through biometric features of measurable body parts of the user, it is necessary to collect biometric data for the user's body parts and extract biometric features from the collected biometric data, so as to identify the user based on the extracted biometric features. For example, if the identity recognition trigger event is to trigger identity recognition through face, the terminal needs to collect facial data from the user's face and identify the user based on the collected facial data, such as a facial image; for another example, if the identity recognition trigger event is to trigger identity recognition through palm, the terminal needs to collect palm data from the user's palm and identify the user based on the collected palm data.

[0068] The target user refers to a user who needs to be identified, and specifically can be a user who triggers an identification event. For example, when a user passes through the access control system, the user can enter the data collection area of ​​the access control system. In the data collection area, when the access control system detects the presence of a user, it indicates that identity identification is required, and then identity identification is triggered. The access control system collects the biometric data of the target user in the data collection area, such as the target user's face data, finger data, or palm data. The target part is the part of the human body corresponding to the biometric data that needs to be collected, and the target part is related to the biometric data or biometric features involved in the identification. For example, if the identification is based on the face, the corresponding target part is the face part of the target user who needs to be identified, and the collected biometric data is the face data, and the biometric features identified by the user are the face features. For another example, if the identification is based on the palm, the corresponding target part is the palm part of the target user who needs to be identified, and the collected biometric data is the palm data, and the biometric features identified by the user are the palm features.

[0069] The local hidden deformation form refers to the transformation of the form of the target part of the target user, thereby partially hiding the global part of the target part, so that the local biological data of the target part can be hidden, avoiding the leakage of the global biological features of the target part during the biometric feature collection process of identity recognition, thereby ensuring the security of the global biological features of the target part. The local hiding of the target part can be flexibly implemented according to actual needs. For example, in addition to directly hiding the target part locally physically, the target part can also be deformed to locally hide the global biological features of the target part. For example, according to the flexibility of the target part itself, the application scenario of identity recognition, etc., the target part can be deformed to make the target part in a locally hidden deformation form. For example, when the target part is a face, the user's face can be partially blocked, such as blocking a specific part of the user's face with an external blocking object, thereby blocking part of the face area; the user can also control the face to transform, such as making a specific expression, so that the face is in a locally hidden deformation form, hiding the local facial features of the face. For another example, when the target part is the palm, considering that the palm part can be flexibly transformed by the fingers, different gestures or hand shapes can be made by the fingers to cover or deform the palm, so that the palm is in a partially hidden deformed form. Specifically, the palm can make an "OK" gesture with the index finger and thumb, and the biometric features of the index finger and thumb in the palm can be hidden, so that the complete biometric features in the palm can be partially hidden. When the target part is in different partially hidden deformed forms, when collecting biometric features of the target part, different biometric features that are partially hidden can be collected, and identity recognition can be performed through different locally hidden biometric features, which can effectively prevent the global biometric features of the target part from being leaked.

[0070] The biometric feature combination is a combination of at least two types of biometric features at the target part. The biometric feature combination is obtained by combining different types of biometric features, specifically by combining at least two types of biometric features. The type of biometric features in the biometric feature combination corresponds to the part type of the target part. For example, when the target part is a face, the biometric features in the biometric feature combination are various types of facial features in the face part, such as bone features, texture features, geometric features, and representational features. The type of facial features is determined according to the facial features involved in the face-based identity recognition process. In a specific application, if the facial features are obtained by different facial feature extraction methods, such as extracting features from a facial image by different face recognition models, the biometric feature combination can be obtained by combining facial features extracted by different facial feature extraction methods. For another example, when the target part is a palm, the biometric features in the biometric feature combination are various types of palm features in the palm part, such as palm print features, palm vein features, etc. The biometric combination is obtained by combining different types of biometrics. When combining, the various types of biometrics can be equivalently combined, that is, the biometric combination can be obtained directly based on the various types of biometrics; the various types of biometrics can also be weighted combined, that is, the corresponding combination weights are assigned to the various types of biometrics, so as to weight the various types of biometrics and further ensure the effectiveness of the biometric combination. In the specific implementation, the corresponding combination weights of the various types of biometrics can be determined according to the historical identification results corresponding to the various biometrics, so as to highlight the important biometrics in the biometric combination.

[0071] Specifically, the identity recognition method is applied to an identity recognition terminal, and the user can trigger identity recognition at the terminal. For example, when the user needs to pass through the access control or needs to make a payment, the user clicks the identity recognition control at the terminal to trigger the identity recognition event. The terminal responds to the identity recognition triggering event of the target user, and collects the biometric feature combination of the target part of the target user when the target part is in a partially hidden deformed form. The biometric feature combination includes at least two types of biometric features at the target part. In specific implementation, the terminal responds to the identity recognition triggering event and can issue an identification prompt to the target user to prompt the target user to move the target part to the data collection area, and at the same time prompt the target user to partially hide the target part so that the target part is in a partially hidden deformed form. The terminal collects biometric data for the target part of the target user in the data collection area that is in a partially hidden deformed form, and extracts at least two biometric features from the collected biometric data to obtain the biometric feature combination when the target part is in a partially hidden deformed form.

[0072] In specific applications, the terminal can collect biometric data and extract at least two biometric features from the collected biometric data; the terminal can also only collect biometric data and send the collected biometric data to the server after encryption, so that the server can extract biometric features from the collected biometric data; in addition, the terminal can also extract part of the biometric features from the collected biometric data, that is, the terminal can extract one or several types of biometric features from the biometric data, and send the biometric data and the extracted biometric features to the server, and the server can further extract biometric features from the biometric data, and finally obtain the biometric feature combination when the target part is in a partially hidden deformed form based on the biometric features extracted by the terminal and the biometric features extracted by the server. In one embodiment, the terminal can determine whether the terminal needs to extract biometric features from the collected biometric data and the number of types of biometric features to be extracted when extracting biometric features based on the current busyness of identity recognition and the busyness of the identity recognition processing of the server, so as to effectively balance the computing resource pressure of the terminal and the server and ensure the normal operation of the identity recognition system.

[0073] Step 204, feature matching is performed between the collected biometric combination and the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by identity registration of the target part of the registered user when the target part of the registered user is in a partially hidden deformed form.

[0074] Among them, the registered biometric pattern is a combination of biometric features obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form. The registered user is a user who performs identity registration through the target part. After the user performs identity registration through the target part, the biometric features of the registered user can be pre-stored, so as to identify the user. Specifically, the pre-stored biometric features are compared with the biometric features collected during the identity recognition process. If the comparison is consistent, it can be considered that the identity recognition is successful, so as to determine the user identity of the user to whom the biometric features collected during the identity recognition process belong. Specifically, during identity registration, the registered user partially hides or deforms the target part, so that the target part is in a partially hidden deformed form, and the registered user is registered based on the target part in a partially hidden deformed form. For example, the combination of biometric features collected corresponding to the target part in a partially hidden deformed form can be determined as the registered biometric pattern corresponding to the registered user, and the registered user is subsequently identified through the registered biometric pattern.

[0075] In a specific application, for the same registered user, multiple registered biometric feature modes can be pre-registered, that is, the registered user can control the target part to be in different deformation forms in partial hiding, and perform identity registration respectively through the target part in different deformation forms in partial hiding, so as to pre-register multiple registered biometric feature modes for the registered user, and the registered user can realize identity recognition processing through any registered biometric feature mode in various registered biometric feature modes. For example, if the target part is a human face, the registered user can register the face form 1, face form 2 and face form 3 in different deformation forms of the face in partial hiding respectively. After the identity registration is successful, the user can perform identity recognition through any face form of face form 1, face form 2 and face form 3. Specifically, the terminal can perform feature matching on the currently collected face form with face form 1, face form 2 and face form 3 respectively. If the currently collected face form successfully matches with any face form of face form 1, face form 2 and face form 3, the user's identity recognition processing can be realized. Furthermore, when the registered user registers multiple registered biometric feature modes, the recognition scene conditions for various registered biometric feature modes can also be set. For example, if the target part is the palm, the registered user can register the palm shape A, palm shape B and palm shape C in different deformed forms with the palm partially hidden. After the identity registration is successful, the user can set the recognition scene conditions for various palm shapes, such as the recognition scene condition for palm shape A is scene A, the recognition scene condition for palm shape B is scene B, and the recognition scene condition for palm shape C is scene C. Then, during identity recognition, the registered user can successfully perform identity recognition through palm shape A in scene A, but cannot successfully perform identity recognition through palm shape A in scene B. Palm shape B corresponding to scene B is required to successfully perform identity recognition processing. The recognition scene conditions can be flexibly set according to actual needs, such as time effectiveness conditions, location effectiveness conditions, trigger scene effectiveness conditions, etc., so as to set the effectiveness conditions for various registered biometric feature modes, further improving the accuracy and security of identity recognition.

[0076] The biometric matching result is the matching result obtained by matching each type of biometric in the currently collected biometric combination of the target user with the corresponding type of biometric in the pre-stored registered biometric pattern. The target user can be identified based on the biometric matching result.

[0077] Specifically, the terminal may pre-store a registered biometric pattern, which is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form. That is, the registered biometric pattern may be an identity credential corresponding to the registered user, and the corresponding registered user may be determined through the registered biometric pattern, thereby realizing identity recognition. The terminal performs feature matching on the collected biometric combination and the registered biometric pattern corresponding to each registered user, respectively, to obtain a biometric matching result. If the biometric matching result is a successful match, it indicates that the pre-registered users include the target user, and the identity of the target user may be determined to be the registered user corresponding to the registered biometric pattern in the successfully matched biometric matching result. In specific applications, the collected biometric combination is feature matched with the pre-stored registered biometric pattern, and the obtained biometric matching result may directly include the matching results corresponding to various types of biometrics, or may include the matching results corresponding to various types of biometrics, and the fused matching result obtained after weighted fusion according to the corresponding feature weights, and the feature weights of each type of biometrics may be pre-set according to the actual application scenario of identity recognition.

[0078] In specific implementation, the terminal can store part of the registered biometric feature pattern, and the terminal matches the collected biometric feature combination with the pre-stored part of the registered biometric feature pattern. If the biometric feature matching result is a match failure, the collected biometric feature combination can be further sent to the server, so that the server can perform identity recognition processing based on the collected biometric feature combination through the complete registered biometric feature pattern. Among them, the partial registered biometric feature pattern stored in the terminal can be the registered biometric feature pattern corresponding to the registered user with a high frequency of identity recognition. Specifically, the identity recognition frequency of each registered user in the complete registered biometric feature pattern can be analyzed, such as determining the identity recognition frequency of each registered user triggering identity recognition in a cycle of one month or one week, and storing the registered users with a higher frequency in the identity recognition frequency, such as the registered users whose identity recognition frequency exceeds the threshold frequency, or the registered users whose identity recognition frequency is in the top 80%, in the terminal locally, so that the terminal can quickly perform identity recognition processing on the registered users with high frequency of identity recognition through the locally stored partial registered biometric feature pattern, thereby ensuring the processing efficiency of identity recognition.

[0079] Step 206: Determine the identity recognition result for the target user based on the biometric matching result.

[0080] Among them, the identity recognition result is used to characterize whether the identity recognition of the target user is successful or not. Specifically, the terminal determines the identity recognition result corresponding to the target user according to the obtained biometric matching result. For example, if the biometric matching result is an unsuccessful match, that is, the currently collected biometric combination of the target user does not match the registered biometric patterns corresponding to various pre-stored registered users, indicating that the target user is not a registered user, that is, the target user has not registered in advance, then the identity recognition result of the target user can be determined to be a recognition failure. If the biometric matching result is a successful match, that is, the currently collected biometric combination of the target user successfully matches the registered biometric pattern of the target registered user among the registered users, then the target user can be determined to be the target registered user, thereby obtaining the identity recognition result that the target user is the target registered user, thereby realizing the identity recognition processing of the target user. In specific applications, if the determined identity recognition result for the target user is a recognition failure, the terminal can send a prompt message of recognition failure to the target user, and specifically, the prompt of identity recognition failure can be given to the target user in a perceptible manner, such as through various methods such as voice messages, interface text messages, etc.

[0081] In the above-mentioned identity recognition method, during the identity recognition process, the biometric combination collected for feature matching is a combination of at least two types of biometrics at the target part of the target user when the target part of the target user is in a partially hidden deformed form, and the pre-stored registered biometric pattern is a biometric combination obtained by identity registration of the target part of the registered user when the target part of the registered user is in a partially hidden deformed form. Identity recognition is performed based on the collected biometric combination and the pre-stored registered biometric pattern, which can prevent the global biometrics of the user's target part from being stolen, thereby improving the security of the user's biometrics during the identity recognition process.

[0082] In one embodiment, the target part is the palm, and the biometric feature combination includes at least the palm print feature and the palm vein feature. Figure 3 As shown, the processing steps of feature matching, that is, feature matching the collected biometric combination with the pre-stored registered biometric pattern to obtain the biometric matching result, include:

[0083] Step 302, determining a pre-stored registration biometric feature pattern; the registration biometric feature pattern at least includes a palm print registration feature and a palm vein registration feature obtained by performing identity registration on the palm of the registered user.

[0084] The target part is the palm, and the biometric feature combination includes at least palm print features and palm vein features at the palm. Palm print refers to the palm image from the end of the finger to the wrist, which includes various features such as main lines, wrinkles, fine textures, ridge ends, bifurcation points, etc. that can be used for identity recognition. Palm print features refer to the features reflected by the texture information of the palm, which can be extracted from the palm image by taking an image of the palm. Different users generally correspond to different palm print features, that is, the palms of different users have different texture features, and the identity recognition processing of different users can be realized based on palm print features. Palm vein refers to the palm vein information image, which is used to reflect the vein image information in the palm of the human body, has the ability of live body recognition, and can be obtained by taking pictures with an infrared camera. Palm vein features are the vein features of the palm part obtained based on palm vein analysis. Different users generally correspond to different palm vein features, that is, the palms of different users have different vein features, and the identity recognition processing of different users can also be realized based on palm vein features. The biometric feature combination is obtained by combining at least the palm print feature and the palm vein feature at the palm, and in specific applications, it can also include various types of palm biometric features such as palm bone features and palm contour features.

[0085] The registered biometric feature pattern is a combination of biometric features reserved when the registered user performs identity registration in advance, which at least includes the palm print registration feature and palm vein registration feature obtained by performing identity registration on the palm of the registered user. The palm print registration feature and palm vein registration feature are the palm print features and palm vein features of the palm reserved by the registered user when registering the identity. Specifically, when the registered user registers the identity through the palm, the palm can be partially hidden in a deformed form, such as the user can make different gestures to change the shape of the palm to achieve partial hiding of the palm, thereby also achieving partial hiding of the palm print features and palm vein features at the palm. The terminal collects palm data of the palm of the registered user in the partially hidden deformed form, thereby obtaining at least the palm print registration feature and palm vein registration feature, and registers the identity of the registered user at least through the palm print registration feature and palm vein registration feature, and obtains the registered biometric feature pattern corresponding to the registered user.

[0086] Specifically, after acquiring the target user's biometric combination, the terminal determines a pre-stored registered biometric pattern, such as determining the pre-stored registered biometric pattern from a local database, or querying the pre-stored registered biometric pattern from a server, so as to perform feature matching through the registered biometric pattern, thereby realizing identity recognition processing. The registered biometric pattern at least includes a palm print registration feature and a palm vein registration feature obtained by performing identity registration on the palm of the registered user.

[0087] In specific applications, the registered biometric feature pattern may also include other types of biometric features at the palm. For example, the registered biometric feature pattern may include palm print registration features, palm vein registration features, contour registration features, bone registration features, etc. If the collected biometric feature combination includes palm print features and palm vein features, the palm print features and palm vein features in the biometric feature combination can be feature matched with the palm print registration features and palm vein registration features in the registered biometric feature pattern to achieve identity recognition; if the collected biometric feature combination includes palm vein features and palm contour features, the palm vein features and palm contour features in the biometric feature combination can be feature matched with the palm vein registration features and contour registration features in the registered biometric feature pattern to achieve identity recognition. The types of biometric features included in the registered biometric feature pattern may be more than the types of biometric features in the biometric feature combination collected during the identity recognition process. For example, if the type of biometric features included in the registered biometric feature pattern is M, and the type of biometric features in the biometric feature combination collected during the identity recognition process is N, then M≥N, so that it can be applicable to scenarios where different types of biometric features are collected during the identity recognition process, expanding the application scenarios of identity recognition.

[0088] Step 304, performing palm print feature matching on the palm print feature in the collected biometric feature combination and the palm print registration feature to obtain a palm print feature matching result.

[0089] Among them, the palmprint feature matching result is the matching result obtained by feature matching based on the palmprint feature, reflecting the recognition result of identity recognition through the palmprint. Specifically, after determining the pre-stored registered biometric feature pattern, the terminal performs palmprint feature matching on the palmprint feature in the collected biometric feature combination and the palmprint registration feature in the registered biometric feature pattern to obtain the palmprint feature matching result. In specific implementation, the palmprint feature matching can be a palmprint feature similarity calculation, so as to obtain a palmprint feature matching result including the palmprint similarity. If the palmprint similarity exceeds the palmprint similarity threshold, it can be considered that the palmprint match is consistent, otherwise it is considered that the palmprint match is inconsistent.

[0090] Step 306 , performing palm vein feature matching on the palm vein feature in the collected biometric feature combination and the palm vein registration feature to obtain a palm vein feature matching result.

[0091] Among them, the palm vein feature matching result is the matching result obtained by feature matching based on the palm vein feature, reflecting the identification result of identity identification through the palm vein. Specifically, after determining the pre-stored registered biometric feature pattern, the terminal performs palm vein feature matching on the palm vein feature in the collected biometric feature combination with the palm vein registration feature in the registered biometric feature pattern to obtain a palm vein feature matching result. In specific implementation, the palm vein feature matching can be a palm vein feature similarity calculation to obtain a palm vein feature matching result including the palm vein similarity. If the palm vein similarity exceeds the palm vein similarity threshold, the palm vein match can be considered consistent, otherwise the palm vein match is considered inconsistent.

[0092] Step 308: Obtain a biometric feature matching result based on the palm print feature matching result and the palm vein feature matching result.

[0093] Among them, the matching results obtained by feature matching each type of biometric feature in the currently collected biometric feature combination of the target user and the corresponding type of biometric feature in the pre-stored registered biometric feature pattern are respectively obtained, and the target user can be identified according to the biometric feature matching results. Specifically, after obtaining the palm print feature matching results and the palm vein feature matching results, the terminal obtains the biometric feature matching results based on the palm print feature matching results and the palm vein feature matching results. For example, the terminal can perform weighted fusion on the palm print feature matching results and the palm vein feature matching results, so as to obtain the biometric feature matching results according to the weighted fusion results.

[0094] In this embodiment, the target part is the hand, and the biometric feature combination includes at least palm print features and palm vein features at the palm. Feature matching is performed through the palm print features and palm vein features at the palm to achieve identity recognition of the target user. In the process of identity recognition processing based on the user's palm, the global palm features of the user's palm part, at least including the palm print features and palm vein features, can be prevented from being stolen, thereby improving the security of the user's palm features during the identity recognition process.

[0095] In one embodiment, obtaining a biometric matching result based on a palm print feature matching result and a palm vein feature matching result includes: determining a palm print matching weight and a palm vein matching weight; and performing weighted fusion on the palm print feature matching result and the palm vein feature matching result based on the palm print matching weight and the palm vein matching weight to obtain a biometric matching result.

[0096] Among them, the palm print matching weight is the weighted weight corresponding to the palm print feature, which reflects the importance of the palm print feature in the identity recognition process. The higher the palm print matching weight, the more important the palm print feature is in the identity recognition process, and the stronger the correlation between the identity recognition result and the palm print feature. The palm vein matching weight is the weighted weight corresponding to the palm vein feature, which is used to characterize the importance of the palm vein feature in the identity recognition process. The larger the palm vein matching weight, the greater the impact of the palm vein feature matching result on the identity recognition result. The palm print matching weight and the palm vein matching weight can be flexibly set according to actual needs, and can be set according to different identity recognition scenarios, or according to the recognition accuracy of the palm print feature and the palm vein feature in the historical identity recognition. In addition, the palm print matching weight and the palm vein matching weight can also be set according to the palm of the registered user in a partially hidden deformed form when performing identity registration. Specifically, the corresponding weight can be set according to the feature richness corresponding to the palm print feature and the palm vein feature collected from the palm during identity registration. For example, when a user is registering his / her identity, palm features are collected for a palm that is partially hidden and deformed. The collected palm vein area is larger and the palm vein features are richer, while the palm print area is smaller and the palm print feature richness is lower. In this case, the palm vein matching weight can be set higher and the palm print matching weight can be set lower, thereby ensuring that accurate identity recognition can be performed through the palm print registration features and palm vein registration features in the registered biometric feature mode.

[0097] Specifically, after obtaining the palm print feature matching result and the palm vein feature matching result, the terminal determines the palm print matching weight and the palm vein matching weight. The palm print matching weight and the palm vein matching weight can be uniformly set, that is, in the identity recognition scenario, for the registered biometric feature patterns corresponding to different registered users, the palm print matching weight and the palm vein matching weight are the same setting value; the palm print matching weight and the palm vein matching weight can also be set separately according to the registered biometric feature patterns corresponding to the registered users, that is, different registered users can be set with different palm print matching weights and palm vein matching weights. For each registered biometric feature pattern corresponding to a registered user, the terminal determines the corresponding palm print matching weight and the palm vein matching weight. Based on the determined palm print matching weight and palm vein matching weight, the terminal performs weighted fusion on the palm print feature matching result and the palm vein feature matching result, such as weighting the palm print feature matching result by the palm print matching weight, and weighting the palm vein feature matching result by the palm vein matching weight, and fusion of the weighted processing results of the two to obtain the biometric matching result.

[0098] In this embodiment, the palm print feature matching results and the palm vein feature matching results are weighted and fused through the preset palm print matching weights and palm vein matching weights. The importance of the palm print features and the palm vein features can be adjusted according to actual needs, thereby improving the accuracy of the obtained biometric feature matching results, which is conducive to improving the accuracy of identity recognition.

[0099] In one embodiment, the target part is a palm; in response to an identity recognition trigger event, for the target part of the target user, a biometric feature combination is collected when the target part is in a partially hidden deformed form, including: in response to the identity recognition trigger event, entering a palm feature collection state; in the palm feature collection state, collecting a biometric feature map of the palm of the target user; when it is determined that the palm of the target user is in a partially hidden palm form, extracting the biometric feature combination of the palm of the target user from the biometric feature map.

[0100] The target part is the palm, and the palm feature collection state is the collection state of biometric data collection and processing for the palm of the target user. The biometric feature map is a feature image obtained by collecting biometric data for the palm, which may specifically include but is not limited to various types of feature images including color images, depth images, infrared images, etc. Based on the biometric feature map, the biometric features corresponding to the corresponding part can be obtained, such as extracting palm print features from the color image of the palm, extracting depth information from the depth image of the palm, extracting palm vein features from the infrared image of the palm, etc. The type of the biometric feature map can correspond to the type of biometric feature to be collected, so that various types of biometric features required for identity recognition can be extracted from the collected biometric feature map.

[0101] Specifically, when the target user triggers identity recognition, the terminal responds to the target user's identity recognition triggering event and enters the palm feature collection state. For example, the terminal can turn on the palm feature collection device, and specifically can turn on various types of cameras to collect palm features of the target user's palm. In the palm feature collection state, the terminal collects the biometric feature map of the target user's palm. In a specific application, when the terminal determines that the palm of the target user is detected, the terminal can collect the feature map for the palm of the target user to obtain the biometric feature map of the target user's palm. For example, if the terminal does not detect the palm of the target user in the palm feature collection state, or the detected part is not the palm part, such as the leg, the terminal can issue a prompt to prompt the target user to cooperate in placing the palm in the collection area so that the terminal can collect the biometric feature map. After obtaining the biometric feature map of the target user's palm, the terminal extracts the biometric feature combination of the target user's palm from the biometric feature map of the target user's palm. Specifically, the terminal can make a rough judgment on the biometric feature map to determine whether the palm of the target user is in a partially hidden palm state. If it is determined based on the biometric feature map that the palm of the target user is not in a partially hidden palm state, that is, the target user has not covered or deformed the palm, and has not partially hidden the global biometric features of the palm, then the terminal can directly identify the target user as a recognition failure. When the palm of the target user is in a partially hidden palm state, the terminal extracts biometric features from the biometric feature map to obtain various types of biometric features at the palm of the target user, thereby obtaining a biometric feature combination of the palm of the target user.

[0102] In specific applications, when the target part is the palm, in addition to pre-storing the registered biometric pattern, the corresponding registered palm features when the palm is not partially hidden can be pre-stored. When the terminal biometric map determines that the target user's palm is not in a partially hidden palm shape, the biometric map can also be used to extract biometric features, and the extracted biometric features can be matched with the registered palm features when the registered user's palm is not partially hidden, thereby realizing the identity recognition of the target user through the complete palm shape. However, there is a security risk that the biometric features of the complete palm may be stolen during the processing.

[0103] In this embodiment, the target part is the palm. When it is determined that the palm of the target user is in a partially hidden palm shape based on the biometric feature map collected in the palm feature collection state, the biometric feature combination of the palm of the target user is extracted from the biometric feature map. Therefore, in the process of identity recognition based on the user's palm, there is no need to collect the global palm features of the user's palm part, which can prevent the global palm features of the user's palm part from being stolen, thereby improving the security of the user's palm features in the identity recognition process.

[0104] In one embodiment, when it is determined that the palm of the target user is in a partially hidden palm shape, a biometric combination of the palm of the target user is extracted from a biometric map, including: performing palm shape recognition on the palm of the target user based on the biometric map to obtain a palm shape recognition result; when the palm shape recognition result indicates that the palm of the target user is in a partially hidden palm shape, extracting at least two types of palm biometrics of the palm of the target user from the biometric map; and obtaining a biometric combination of the palm of the target user based on the at least two types of palm biometrics.

[0105] Among them, palm shape recognition is to identify the palm shape of the target user to determine whether the palm of the target user is in a partially hidden deformed form, that is, whether the palm of the target user is partially hidden. The palm shape recognition result is used to characterize the palm shape corresponding to the palm of the target user collected by the terminal, such as a partially hidden deformed form, or a complete form that is not partially hidden. Palm biometrics are biometrics corresponding to the palm part, such as palm print features, palm vein features, bone features, palm contour features, etc.

[0106] Specifically, after obtaining the biometric map of the palm of the target user, the terminal performs palm morphology recognition on the palm of the target user based on the biometric map, such as performing palm morphology recognition on the integrity of the fingers, the positions of the fingers, the gestures reflected by the palm, etc. in the biometric map, to obtain a palm morphology recognition result. If the palm morphology recognition result indicates that the palm of the target user is in a partially hidden palm shape, indicating that the target user has partially hidden the palm, the terminal extracts at least two types of palm biometric features of the palm of the target user from the biometric map, such as extracting at least palm print features and palm vein features from the biometric map. The terminal obtains a biometric feature combination of the palm of the target user based on the extracted at least two types of palm biometric features, such as the terminal can obtain a biometric feature combination of the palm of the target user based on the extracted palm print features and palm vein features.

[0107] In this embodiment, the target part is the palm. When it is determined that the palm of the target user is in a partially hidden palm shape based on the palm shape recognition result corresponding to the biometric feature map collected in the palm feature collection state, at least two types of palm biometric features are extracted from the biometric feature map, and a biometric feature combination of the palm of the target user is obtained based on the at least two types of palm biometric features. Therefore, in the process of identity recognition processing based on the user's palm, there is no need to collect the global palm features of the user's palm part, which can prevent the global palm features of the user's palm part from being stolen, thereby improving the security of the user's palm features in the identity recognition process.

[0108] In one embodiment, the identity recognition method further includes: determining a target gesture corresponding to the palm of the target user according to the biometric feature map; performing gesture recognition on the target gesture based on each registered gesture in a local registered gesture library to obtain a gesture recognition result.

[0109] Among them, the biometric feature map is a feature image obtained by collecting biometric data for the palm. Based on the biometric feature map, the biometric features corresponding to the corresponding part can be obtained. The target gesture refers to the gesture presented by the palm of the target user collected. When the target part is the palm, the palm is partially hidden. The user can make different gestures with the hand to partially hide the palm, so as to facilitate the user to partially hide the palm part. The registered gesture refers to the gesture presented by the registered user's hand when the registered user performs identity recognition registration through the palm and controls the palm to be in a deformed form of partial hiding. The registered gestures corresponding to each registered user are gathered to obtain a registered gesture library. The gesture recognition result is the recognition result of the target gesture of the target user based on each registered gesture in the registered gesture library. Through the gesture recognition result, it can be determined whether the target gesture of the target user is consistent with the registered gesture corresponding to the registered user. If the registered gestures corresponding to the target user and each registered user in the registered gesture library are inconsistent, it can be determined from the gesture level that the target user cannot be successfully matched with the registered user, and it can be determined that the identity recognition of the target user has failed.

[0110] Specifically, after obtaining the biometric map corresponding to the palm of the target user, the terminal determines the target gesture corresponding to the palm of the target user based on the biometric map. For example, the terminal can perform gesture recognition based on the biometric map to determine the target gesture corresponding to the palm of the target user. The terminal queries the registered gesture library pre-stored on the local side, and the registered gesture library includes the registered gestures corresponding to the registered user when registering his identity. The terminal performs gesture recognition on the target gesture based on each registered gesture in the registered gesture library on the local side to obtain a gesture recognition result. Specifically, the terminal can perform gesture matching on the target gesture of the target user with each registered gesture in the registered gesture library to obtain a gesture recognition result.

[0111] Furthermore, the collected biometric combination is feature matched with the pre-stored registered biometric pattern to obtain a biometric matching result, including: when the gesture recognition result is recognition passed, the collected biometric combination is sent to the server, so that the server performs feature matching on the collected biometric combination and the pre-stored registered biometric pattern; and the biometric matching result returned by the server is obtained.

[0112] Specifically, after obtaining the gesture recognition result, if the gesture recognition result is recognition passed, it indicates that the target gesture corresponding to the palm of the target user matches the registered gesture of the registered user, and at the gesture level, the target user can realize identity recognition, then the terminal sends the collected biometric combination to the server, so that the server performs feature matching on the collected biometric combination and the pre-stored registered biometric pattern, and the terminal receives the biometric matching result returned by the server, so that the identity recognition result for the target user can be determined based on the biometric matching result, and the identity recognition processing of the target user can be realized. When the gesture recognition result corresponding to the target gesture of the target user is recognition passed, it indicates that the target user performs identity pre-recognition processing by using gestures, and the terminal sends the collected biometric combination to the server, so that the server performs feature matching on the collected biometric combination and the pre-stored registered biometric pattern, so that the server performs identity recognition from the biometric combination level, and realizes secondary processing of the target user's identity recognition, which can improve the processing efficiency of identity recognition under the premise of ensuring the accuracy of identity recognition.

[0113] In this embodiment, the terminal performs gesture recognition on the target gesture corresponding to the palm of the target user through the registered gestures corresponding to each registered user in the local registered gesture library, thereby pre-identifying the target user from the gesture level. After the identity pre-identification is passed, the server performs feature matching on the collected biometric combination to achieve identity recognition at the biometric combination level. Gesture recognition can be used to pre-identify the identity by the terminal, which can improve the processing efficiency of identity recognition, and the terminal does not need to store the registered biometric pattern, which can ensure the security of the registered biometric pattern.

[0114] In one embodiment, the identity recognition method also includes: responding to a trigger event of registered identity recognition, entering a biometric collection state; in the biometric collection state, collecting a biometric combination to be registered when the target part of the registered user is in a partially hidden deformed form; when the biometric combination to be registered meets the identity recognition registration conditions, performing identity registration on the target part of the registered user based on the biometric combination to be registered to obtain a registered biometric pattern.

[0115] Among them, the triggering event of registered identity recognition refers to the event that triggers identity recognition registration, which may specifically include but is not limited to operations and instructions that trigger identity recognition registration. The biometric collection state is a collection state for collecting and processing biometric data for the target part of the target user. For example, the terminal can start the biometric collection device, and specifically various types of cameras can be turned on to collect biometrics for the target part of the target user. In the biometric collection state, the terminal can collect the biometric map of the target part of the target user. The identity recognition registration condition is a restriction condition for the biometric combination to be registered. Only when the biometric combination to be registered meets the identity recognition registration condition can the identity recognition registration be performed for the registered user through the biometric combination to be registered, thereby obtaining the registered biometric pattern corresponding to the registered user.

[0116] Specifically, when the registered user registers his identity through the target part, the registered user can trigger a trigger event of the registered identity recognition, and the terminal enters the biometric feature collection state in response to the trigger event of the registered identity recognition. In the biometric feature collection state, the terminal collects the biometric feature combination to be registered when the target part of the registered user is in a partially hidden deformed form. Specifically, the terminal can send a prompt message to prompt the registered user to make the target part in a partially hidden deformed form. The terminal collects the biometric feature map corresponding to the target part in the partially hidden deformed form, and extracts at least two types of biometric features from the biometric feature map. The terminal obtains the biometric feature combination to be registered based on the obtained at least two types of biometric features.

[0117] The terminal obtains the preset identity recognition registration conditions, which may specifically include the validity conditions of the biometric combination. The terminal compares the biometric combination to be registered with the identity recognition registration conditions. For example, the terminal may analyze the validity of the biometric combination to be registered to determine whether the biometric combination to be registered can uniquely distinguish the registered user, obtain the validity score, and compare the validity score of the biometric combination to be registered with the validity score threshold in the identity recognition registration conditions. If the validity score of the biometric combination to be registered is greater than the validity score threshold, it is considered that the biometric combination to be registered meets the identity recognition registration conditions. The terminal performs identity registration on the target part of the registered user based on the biometric combination to be registered, thereby obtaining the registered biometric pattern corresponding to the registered user, and the terminal can perform user identity recognition processing through the registered biometric pattern.

[0118] In this embodiment, when the registered user performs identity registration through the target part, when the terminal determines that the obtained biometric feature combination to be registered meets the identity recognition registration condition, the terminal performs identity registration on the target part of the registered user based on the biometric feature combination to be registered to obtain the registered biometric feature pattern. The identity recognition registration condition ensures the validity of the biometric feature combination to be registered when performing identity registration for the registered user, and the validity of the registered biometric feature pattern can be ensured, so that accurate identity recognition processing can be performed through the registered biometric feature pattern.

[0119] In one embodiment, the identity recognition method also includes: in response to a triggering event of replacing a registered biometric pattern, collecting a biometric combination to be replaced when the target part of the registered user is in a partially hidden deformed form; performing feature validity analysis on various types of biometrics in the biometric combination to be replaced to obtain a feature analysis result; when the feature analysis result indicates that the biometric combination to be replaced is valid, replacing the biometric combination in the registered biometric pattern with the biometric combination to be replaced.

[0120] The triggering event for replacing the registered biometric pattern refers to an event that triggers the replacement of the registered biometric pattern, which may specifically include but is not limited to operations and instructions that trigger the replacement of the registered biometric pattern. The biometric combination to be replaced is a biometric combination obtained by collecting biometric features from the target part of the registered user after the event of triggering the replacement of the registered biometric pattern. Feature validity analysis refers to analyzing the validity of various types of biometrics in the biometric combination to be replaced to determine whether the various types of biometrics in the biometric combination to be replaced can uniquely distinguish the registered user. If the biometric combination to be replaced can uniquely distinguish the registered user, it means that the biometric combination to be replaced can be used to identify the registered user, that is, it can be considered that the various types of biometrics in the biometric combination to be replaced are valid, and the biometric combination to be replaced can be used for identity recognition processing.

[0121] Specifically, when the registered user needs to replace the pre-stored registered biometric feature pattern, the registered user can trigger a trigger event for replacing the registered biometric feature pattern, and the terminal enters a biometric feature collection state in response to the trigger event for replacing the registered biometric feature pattern. In the biometric feature collection state, the terminal collects the biometric feature combination to be replaced when the target part of the registered user is in a partially hidden deformed form. Specifically, the terminal can send a prompt message to prompt the registered user to make the target part in a partially hidden deformed form. The terminal collects the biometric feature map corresponding to the target part in the partially hidden deformed form, and extracts at least two types of biometric features from the biometric feature map. The terminal obtains the biometric feature combination to be replaced based on the at least two types of biometric features obtained. The terminal performs feature validity analysis on various types of biometric features in the biometric feature combination to be replaced to determine whether the biometric feature combination to be replaced can uniquely distinguish the registered user, and obtain a feature analysis result. In specific applications, the terminal can analyze the uniqueness of various types of biometric features in the biometric feature combination to be replaced to determine whether the biometric feature combination to be replaced has the unique characteristics of the registered user and whether it can uniquely distinguish the registered user. For example, the terminal can use various types of biometric features in the biometric feature combination to be replaced for identity recognition. If other users can be recognized, it means that the biometric feature combination to be replaced cannot uniquely distinguish the registered user. If other users cannot be recognized, the registered user can be uniquely distinguished by the biometric feature combination to be replaced. When the obtained feature analysis result indicates that the biometric feature combination to be replaced is valid, that is, the registered user can be uniquely distinguished by the biometric feature combination to be replaced, the terminal replaces the biometric feature combination in the registered biometric feature pattern with the biometric feature combination to be replaced, thereby realizing the replacement processing of the registered biometric feature pattern, and the terminal can perform user identity recognition processing by the replaced registered biometric feature pattern.

[0122] In this embodiment, when the registered user needs to replace the pre-stored registered biometric pattern, when the terminal determines that the obtained feature analysis result corresponding to the biometric combination to be replaced indicates that the biometric combination to be replaced is valid, the terminal replaces the biometric combination in the registered biometric pattern of the registered user based on the biometric combination to be replaced, thereby realizing the replacement processing of the registered biometric pattern.

[0123] In one embodiment, the identity identification method further includes: determining resource transfer parameters in response to a resource transfer triggering event; determining a target resource account associated with the target user according to the identity identification result; and performing resource transfer to the target resource account based on the resource transfer parameters.

[0124] Among them, resources are assets that can be exchanged for subject matter. Resources can be funds, electronic vouchers, shopping vouchers and virtual red envelopes, etc. Virtual red envelopes are virtual objects with certain fund numerical attributes. For example, funds can be exchanged for goods of equal value after a transaction. Resource transfer refers to the exchange of resources, including resource transferees and resource transferees. Resources are transferred from resource transferees to resource transferees. For example, in the payment process of shopping, funds are transferred as resources. Resource transfer trigger events refer to events that trigger resource transfers, which may specifically include but are not limited to operations and instructions that trigger resource transfers. Resource transfer trigger events can be triggered by users who need to perform resource transfer processing, such as by resource transferees in the resource transfer processing, or by resource transferees in the resource transfer processing. Resource transfer is to transfer a certain amount of resources held by resource transferees to resource transferees. Resource transfer trigger events can be flexibly set according to actual needs. Resource transfer parameters are related parameters for resource transfer processing to be performed corresponding to resource transfer trigger events, and may specifically include but are not limited to various parameter information including resource transferee, resource transferor, resource transfer amount, discount amount, order number, resource transfer time, resource transfer terminal, etc. The target resource account is the resource account associated with the target user. By performing resource transfer operations on the target resource account, the resource transfer processing of the target user can be realized.

[0125] Specifically, the terminal can determine resource transfer parameters in response to a resource transfer trigger event, such as determining a resource transfer amount, a resource transferee, etc. If the user identity corresponding to the target user can be determined based on the identity recognition result, the terminal can determine the target resource account associated with the target user based on the identity recognition result. Specifically, the terminal can determine the user identity corresponding to the target user based on the identity recognition result, and determine the target resource account associated with the target user based on the user identity corresponding to the target user, and the target resource account includes the resources of the target user. The terminal transfers resources to the target resource account based on the determined resource transfer parameters, such as transferring the resources in the target resource account to the resource transferee in the resource transfer parameters according to the resource transfer amount in the resource transfer parameters, thereby realizing resource transfer processing for the target user. In a specific implementation, the maximum transfer amount supported during the resource transfer process is related to the locally hidden deformation form of the target part of the target user. For example, when the target part is in the partially hidden deformation form 1, the maximum supported transfer amount is a; when the target part is in the partially hidden deformation form 2, the maximum supported transfer amount is b; when the target part is in the partially hidden deformation form 3, the maximum supported transfer amount is c, so that the target user can further realize resource transfer processing with different transfer amounts by being in different locally hidden deformation forms.

[0126] In this embodiment, the target resource account associated with the target user is determined based on the identity recognition result, and when a resource transfer triggering event is triggered, resource transfer processing is performed according to corresponding resource transfer parameters through the target resource account associated with the target user, thereby realizing resource transfer processing through the identification of the target part, while improving the processing efficiency of resource transfer and ensuring the security of the user's complete biometric characteristics during the resource transfer process.

[0127] In one embodiment, feature matching is performed on the collected biometric combination with the pre-stored registered biometric pattern to obtain a biometric matching result, including: determining the pre-stored registered biometric pattern; feature matching is performed on each type of biometric in the collected biometric combination with the registered biometric of the corresponding type of each type of biometric in the registered biometric pattern to obtain feature matching results corresponding to each type of biometric; and the biometric matching result is obtained based on the feature matching results corresponding to each type of biometric.

[0128] The registered biometric pattern is a combination of biometric features obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form. The registered biometric pattern includes at least two types of registered biometric features at the target part. The registered biometric feature is a biometric feature collected by the terminal from the target part of the registered user in a partially hidden deformed form when the registered user performs identity registration.

[0129] Specifically, when performing feature matching on the collected biometric feature combination, the terminal determines the pre-stored registered biometric feature pattern, and performs feature matching on each type of biometric feature in the collected biometric feature combination with the registered biometric feature of the corresponding type of each type of biometric feature in the registered biometric feature pattern, and obtains the feature matching results corresponding to each type of biometric feature. Each type of biometric feature in the biometric feature combination corresponds to a feature matching result, and the terminal obtains the biometric feature matching result based on the feature matching results corresponding to each type of biometric feature. Specifically, the terminal can synthesize the feature matching results corresponding to each type of biometric feature, such as the terminal can weightedly fuse the corresponding feature matching results according to the weights corresponding to each type of biometric feature to obtain the biometric feature matching result.

[0130] In this embodiment, each type of biometric in the collected biometric combination is feature matched with the registered biometric of the corresponding type of each type of biometric in the registered biometric pattern, and a biometric matching result is obtained based on the feature matching results corresponding to each type of biometric. By achieving feature matching of multiple types of biometrics in the biometric combination, the accuracy of identity recognition can be ensured.

[0131] In one embodiment, feature matching is performed on the collected biometric combination with a pre-stored registered biometric pattern to obtain a biometric matching result, including: determining a triggering scenario corresponding to an identity recognition triggering event; determining a target registered biometric pattern associated with the triggering scenario from the pre-stored registered biometric pattern; and feature matching is performed on the collected biometric combination with the target registered biometric pattern to obtain a biometric matching result.

[0132] Among them, the triggering scene is the scene corresponding to the identity recognition triggering event, such as the access control scene, payment scene, game anti-addiction scene and other scenes. Different registered biometric patterns can be set for different triggering scenes. For example, the registered biometric pattern 1 can be set in the access control scene, and the registered biometric pattern 2 can be set in the payment scene, so that the user can perform identity recognition in the access control scene through the registered biometric pattern 1, and perform identity recognition in the payment scene through the registered biometric pattern 2. The target registered biometric pattern is the registered biometric pattern associated with the triggering scene, that is, the identity recognition processing is realized in the triggering scene through the target registered biometric pattern.

[0133] Specifically, when performing feature matching on the collected biometric combination, the terminal determines the triggering scenario corresponding to the identity recognition triggering event, such as determining the triggering time, triggering location, triggering device, triggering method, etc. of the identity recognition triggering event, thereby determining the triggering scenario corresponding to the identity recognition triggering event. The triggering scenario may include different triggering times, triggering locations, or triggering environments, etc., and corresponding registered biometric patterns are set for different triggering scenarios to further ensure the accuracy of identity recognition. The terminal determines the target registered biometric pattern associated with the triggering scenario from the pre-stored registered biometric patterns, and the target registered biometric pattern is suitable for performing identity recognition processing on the user under the triggering scenario. The terminal performs feature matching on the collected biometric combination with the target registered biometric pattern, and specifically, each type of biometric in the collected biometric combination may be feature matched with the registered biometric of the corresponding type of each type of biometric in the target registered biometric pattern, and obtains the biometric matching result according to the feature matching results corresponding to each type of biometric.

[0134] In this embodiment, feature matching is performed on the collected biometric combination according to the target registered biometric pattern associated with the trigger scenario corresponding to the identity recognition trigger event to realize identity recognition processing under the trigger scenario. Different trigger scenarios can correspond to different target registered biometric patterns, which can ensure the accuracy of identity recognition.

[0135] In one embodiment, the registered biometric pattern includes at least two biometric combinations obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in different deformed forms of partial concealment. The collected biometric combination is feature matched with the pre-stored registered biometric pattern to obtain a biometric matching result, including: the collected biometric combination is feature matched with various biometric combinations in the pre-stored registered biometric pattern to obtain a biometric matching result corresponding to each biometric combination in the registered biometric pattern.

[0136] Among them, the registered biometric pattern includes at least two biometric combinations obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in different deformation forms of local concealment. The registered biometric pattern includes multiple biometric combinations, and each biometric combination is obtained by combining the corresponding collected biometrics when the target part of the registered user is in different deformation forms of local concealment. Specifically, when performing feature matching on the collected biometric combination, the terminal performs feature matching on the collected biometric combination with various biometric combinations in the pre-stored registered biometric pattern, and obtains the biometric matching results corresponding to each biometric combination in the registered biometric pattern. Specifically, the terminal performs feature matching on the collected biometric combination with various biometric combinations in the registered biometric pattern, and obtains the biometric matching results corresponding to each biometric matching.

[0137] Furthermore, the identity recognition result for the target user is determined based on the biometric matching result, including: determining the identity recognition result for the target user based on the successfully matched biometric matching results among the biometric matching results corresponding to the various biometric combinations.

[0138] Specifically, the terminal determines a successfully matched biometric matching result from the biometric matching results corresponding to each of the various biometric matching results, and determines the identity recognition result for the target user based on the successfully matched biometric matching result. In specific implementation, as long as there is any successfully matched biometric matching result among the biometric matching results corresponding to each of the various biometric matching results, it can be considered that the match is successful, and the identity recognition result for the target user is determined based on the successfully matched biometric matching result.

[0139] In this embodiment, registered users can pre-register multiple biometric combinations, and users can use any biometric combination to achieve identity recognition processing, so that users can have multiple identity recognition credentials, and identity recognition can be achieved through any identity recognition credential, which can expand the applicable scenarios of identity recognition and help improve the processing efficiency of identity recognition.

[0140] In one embodiment, the identity recognition method also includes: in response to an identity recognition trigger event, collecting a feature image of a target part of a target user; performing a local hiding analysis on the target part based on the feature image to obtain a local hiding analysis result; when the local hiding analysis result indicates that the target part is not in a locally hidden deformed form, determining that the identity recognition result of the target user is an identity recognition failure.

[0141] The identity recognition trigger event refers to an event that triggers identity recognition, and specifically may include but is not limited to operations and instructions that trigger identity recognition, etc. The feature image is an image obtained by collecting features of a target part of a target user, and specifically may be a biometric feature image.

[0142] Specifically, in response to the target user's identity recognition triggering event, the terminal collects biometric features for the target part of the target user, and obtains a feature image of the target part of the target user, such as a color image or an infrared image of the target user's palm. The terminal performs a local hiding analysis on the target part based on the collected feature image to determine whether the target part is in a deformed form of local hiding, and obtains a local hiding analysis result. When the local hiding analysis result indicates that the target part is not in a deformed form of local hiding, it indicates that the target user has not locally hidden the target part, and the terminal can directly determine that the target user's identity recognition result is an identity recognition failure. In a specific implementation, the registered biometric feature pattern pre-stored in the terminal is a biometric feature combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a deformed form of local hiding, and the registered biometric feature pattern corresponds to the target part in a deformed form of local hiding. If the collected target part of the target user is not in a deformed form of local hiding, the identity recognition of the target user can be directly determined, thereby improving the processing efficiency of identity recognition.

[0143] In this embodiment, when it is determined that the target part of the target user is not in a partially hidden deformed form, the terminal directly determines that the identity recognition result of the target user is an identity recognition failure, thereby accurately recognizing the target user based on the part form of the target part of the target user, thereby improving the processing efficiency of identity recognition.

[0144] In one embodiment, the identity recognition method also includes: in response to an identity recognition trigger event, collecting a depth image corresponding to a target part of the target user; performing liveness detection on the target user based on the depth image to obtain a liveness detection result; when the liveness detection result is a detection failure, determining that the identity recognition result of the target user is an identity recognition failure.

[0145] Among them, the identity recognition trigger event refers to the event that triggers identity recognition, which may include but is not limited to operations and instructions that trigger identity recognition. A depth image is an image that includes depth information, and the depth information is the distance between the photographed object and the camera. The depth image can be acquired through a depth camera. Liveness detection is a method of determining the true physiological characteristics of an object, which can effectively resist common attack methods such as photos, face swaps, masks, occlusions, and screen reshoots, thereby helping users identify fraudulent behavior and protect their interests.

[0146] Specifically, the terminal responds to the target user's identity recognition trigger event and collects the depth image corresponding to the target part of the target user. Specifically, the terminal can collect the depth image of the target part of the target user through the depth camera. The terminal performs liveness detection on the target user based on the collected depth image to obtain a liveness detection result. If the liveness detection result is a detection failure, it indicates that the target user currently being detected may be a prosthesis, and the terminal can directly determine that the target user's identity recognition result is an identity recognition failure. If the liveness detection result is a detection pass, the terminal can further perform a step of feature matching the collected biometric combination with the pre-stored registered biometric pattern to obtain a biometric matching result, thereby realizing the identity recognition processing of the target user.

[0147] In this embodiment, liveness detection is performed through a depth image corresponding to a target part of the target user. When the target user is determined to be a real user, identity recognition processing is performed on the target user. When the target user is determined to be a false user, the identity recognition result of the target user is directly determined to be an identity recognition failure, thereby ensuring the accuracy and security of the identity recognition processing.

[0148] The present application also provides an application scenario, which applies the above-mentioned identity recognition method. Specifically, the application of the identity recognition method in the application scenario is as follows:

[0149] A camera is provided in the access control system, which can collect face data of the user's face. When the target user needs to pass through the access control system, an identity recognition trigger event is triggered. The access control terminal of the access control system collects face data of the target user in a partially hidden deformed form, and extracts face features from the collected face data to obtain a biometric combination corresponding to the target user, which includes at least two types of biometric features on the face of the target user. The terminal matches the collected biometric combination with the pre-stored registered biometric pattern to obtain a biometric matching result. The registered biometric pattern is a biometric combination obtained by performing identity registration on the face of the registered user when the face of the registered user is in a partially hidden deformed form. The terminal determines the identity recognition result for the target user based on the biometric matching result. When the target user is determined to be an authorized user of the access control system based on the identity recognition result, the access control system cancels the interception of the target user so that the target user can pass through the access control system.

[0150] The present application also provides an application scenario, which applies the above-mentioned identity recognition method.

[0151] Specifically, the application of the identity recognition method in this application scenario is as follows:

[0152] Palm scanning refers to a method of identifying an individual through biometric features of the palm, such as palm print features and palm vein features. Information such as palm prints and palm veins on a person's palm is similar to that of a person's face. They are very important biometric features and are difficult to change. Therefore, using a person's palm to identify an individual and further transfer resources, such as shopping payments, is one of the trends in identity recognition research. However, if the information on a person's palm is lost or forged by others, the consequences will be serious, which is equivalent to handing over the key to someone else, posing a huge security risk. The biometric features on a person's palm are different from those on passwords, cards, and other identification methods, and are difficult to modify. The palm of a person has rich human feature information. By combining palm print features and palm vein features, human identity recognition and liveness recognition can be achieved, and difficult-to-recognize situations such as twins (face scanning problems) can be avoided. A system that confirms individual identity and liveness features through palm vein and palm print fusion is referred to as a palm vein and palm print multimodal recognition system. In the palm vein print multimodal recognition system, the palm vein print feature is a numerical value or mark extracted from the palm vein print sample for comparison, referred to as the palm feature; the palm vein print sample is an analog or digital representation of the palm vein print image before the palm vein print feature item is extracted, referred to as the palm sample; the palm vein print reference is one or more stored palm vein print sample features belonging to the user and used as the palm vein print comparison object, referred to as the palm reference; the palm vein print probe is a palm sample or feature input to the algorithm for comparison with the palm reference, referred to as the palm probe. Palm vein print verification is one of the palm vein print fusion applications. The generated palm vein standing feature is compared with the stored user's palm vein feature reference given by the user identification information (1:1 comparison) to confirm whether the user is the claimed identity, referred to as palm comparison. Palm vein and palm print recognition refers to one of the palm vein and palm print fusion applications. The generated palm vein and palm print features are compared with all the palm vein and palm print features stored in a specified range (1:N comparison) to determine the user's identity, referred to as palm recognition. Through the palm vein and palm print multimodal recognition system, the palm vein and palm print features can be used to realize the authentication and recognition of the user's identity.

[0153] At the hardware level of the palm scanning device, an RGB or monochrome camera or an IR camera (infrared camera) can be used to collect palm print information. Through the IR image under 940nm infrared fill light, the blood vessel image information in the palm, that is, the palm vein, can be collected, and then assisted by blue (white) fill light, etc., to achieve a better collection effect. At the algorithm level, various methods of feature pattern recognition and AI (Artificial Intelligence) machine learning can be used to perform identity recognition based on palm print features and palm vein features. Among them, the palm print (Palm Print) refers to the texture image information of the palm, which can be obtained by taking pictures with an RGB camera. For example Figure 4As shown in FIG. 1 , a schematic diagram of a palm print is shown in an embodiment. The palm print reflects the information characteristics of the palm surface texture, which can be used to distinguish different palms, thereby realizing identity recognition. The palm vein (Palm Vein) refers to the palm vein information image, which is used to reflect the vein image information in the palm of the human body, has the ability of living body recognition, and can be obtained by taking pictures with an infrared camera. Figure 5 FIG. 1 is a schematic diagram of palm veins in an embodiment. The palm veins reflect the information characteristics of the palm veins and can be used to distinguish different palms, thereby realizing identity recognition.

[0154] Palm scanning has many advantages and is easy for people to use. It is gradually becoming popular in various application fields. At present, during the data collection process, the default requirement is that the palm of the person being collected should be open and not covered. Figure 6 As shown in FIG. 1 , a palm scanning device is used. The palm scanning device is set vertically. The user opens his palm and the palm scanning device collects biometric features of the palm to realize identity recognition. Figure 7 The figure shows another way to use the palm scanning device. The palm scanning device is set horizontally. The user opens his palm and the palm scanning device collects biometric features of the palm to achieve identity recognition. The features of a person's palm are fixed. If the information of a person's palm is leaked, an attacker can forge a prosthesis with the same information and may forge an identity, causing losses. Even if the live features of the palm vein image are relatively obvious, it is still possible to be hacked, and there is a security risk of leaking the complete features of the user's palm.

[0155] Specifically, when a user actively initiates palm registration on a payment device, the user can use a special hand shape, such as concave or partially blocked, so that the palm features are different from the normal fully open palm, and the palm information is partially hidden or deformed, which can prevent the complete biometric features on the palm from being stolen. There are a large number of biometric feature points on the human palm. By fusing the palm print and palm vein dual factors for identity recognition, rich feature information can still be used for identification after the palm is partially blocked. At the same time, because the hand has fingers and the hand is flexible, it is easier to make a variety of gestures, and the shape made each time is unified, so the blocked part is unified. When others steal palm prints, they usually take a photo of the open and unobstructed palm. Even if the complete palm print information is obtained, it is inconsistent with the user's registration, and it cannot be forged for identification. Furthermore, the user can make several blocking or concave gestures so that he has multiple palm print keys. Therefore, even if the current palm shape pattern1 is stolen, the user can modify the base library again to generate a new pattern2, and the palm swipe can still be used, and it will not be affected by the loss of the original pattern1.

[0156] Furthermore, when partially hiding the palm, specifically forming various hand shapes pattern_shelter by blocking, taking the left hand as an example, Figure 8 As shown, the middle finger, ring finger and little finger can be bent toward the palm side to cover the upper right part of the palm; Fig. 9 As shown, the left side of the palm and the middle of the palm can be partially blocked by bending the thumb toward the palm side; Fig.10 As shown in the figure, the upper left part of the palm can be partially blocked by bending the index finger toward the palm side; Fig.11 As shown in the figure, the upper and middle parts of the palm can be partially blocked by bending the index finger, middle finger, ring finger and little finger toward the palm. When the partially hidden hand shape pattern_Transform is formed by deformation, take the left hand as an example, Fig.12 As shown in , the palm can be sunken inward to form a sunken hand shape, thereby partially hiding the middle of the palm. In addition, the local occlusion and deformation methods can be combined to form the corresponding hand shape pattern_shelter_transform. Taking the left hand as an example, Fig.13 As shown, the upper left part of the palm can be partially hidden by using the orchid finger hand shape.

[0157] In the specific implementation, Fig.14 As shown in the figure, when various hand shape patterns are registered in the base database, the registered user can trigger the activation of palm scanning, and the terminal displays the user agreement. When the registered user confirms the user agreement and agrees to activate palm scanning, the specific hand shape pattern 1 used by the user is entered and pattern 1 is generated in the base database, thereby realizing the palm registration of the registered user. The hand shape pattern can refer to a palm state corresponding to a specific combination of palm prints and palm veins. When performing palm scanning, such as Fig.15 As shown, the user triggers the palm swiping payment, the user swipes his palm, and the terminal collects the user's palm information pattern N. The terminal performs liveness detection based on the user's palm information pattern N. Liveness detection refers to the measurement and analysis of anatomical features, unconscious or conscious reactions, to determine whether the collected palm sample is from a living person at the collection end. If the liveness detection passes, pattern N is matched with various hand shape patterns in the base database. If the match is successful, the user's identity information is displayed, and payment is made through the user's corresponding account. If the liveness detection fails, or the user's identity is not recognized, the user is prompted to swipe his palm again. When the user needs to update the hand shape registered in the base database, such as Fig.16As shown, the user can trigger the update of the palm swiping base database information, and the terminal can display the user agreement. When the user agrees to update the palm swiping base database information, the specific hand pattern2 used by the user is entered and pattern2 is generated in the base database, thereby realizing the palm update of the registered user. The updated pattern2 can replace the registered pattern1, so that the user can perform palm swiping processing through the updated pattern2.

[0158] Specifically, at the hardware level, a large FOV (Field of view) needs to be set to capture a wider range (i.e., higher up and down, wider left and right) to obtain more information and ensure that palm information is not missed. Fig.17 As shown in the figure, it is a schematic diagram of the FOV field of view. The FOV field of view is used to describe the angle range of a given scene imaged by a camera. There are three main types of field of view: horizontal field of view, vertical field of view, and diagonal field of view. Among them, setting a larger FOV can make the diagonal range, vertical range, and horizontal range of the camera larger, thereby ensuring that the palm information is not missed by the camera. Furthermore, as Fig.18 As shown, there is an allowable confusion circle before and after the focus, and the distance between the two confusion circles is called the depth of field, that is, the image still has a clear range before and after the subject (focus point), which is the depth of field. In other words, the front and back depth of the subject, the image blur presented on the film surface, are all within the limited range of the allowable confusion circle. The depth of field varies with the focal length, aperture value, and shooting distance of the lens. For a fixed focal length and shooting distance, the smaller the aperture, the greater the depth of field. When shooting the palm, you can also adjust to a suitable depth of field to ensure that the captured palm image can be used for identity recognition. At the hardware level, a partially occluded palm cannot be judged as an invalid palm. Moreover, although partially occluded, most of the biometric features are still leaked. It is necessary to set up a perfect multi-modality, set up pattern recognition and training of biometric features under various hand shapes, and do a good job of threshold control of algorithm thresholds, FAR (False Accept Rate), FRR (False Reject Rate) and other indicators to ensure accurate identity recognition processing for the palm. Among them, the threshold refers to the boundary value (or value set) based on which the judgment of whether two or more palm vein and palm print samples are from the same person is made. FAR refers to the ratio of the number of false acceptances to the sum of the number of correct rejections and the number of false acceptances, expressed as a percentage. The false acceptance rate is sometimes also called the false matching rate, false recognition rate, false recognition rate, false alarm rate, false positive rate, etc.; FRR refers to the ratio of the number of false rejections to the sum of the number of correct acceptances and false rejections, expressed as a percentage. The false rejection rate is sometimes also called the true rejection rate, false negative rate, false alarm rate, false negative rate, etc.

[0159] In specific palm payment scenarios, such as Fig.19 As shown, in the traditional palm payment, the user opens his palm in front of the palm swiping device, and the palm swiping device collects the user's complete palm features for identity recognition. After the user's identity is determined, the payment is processed for the user. The complete information of the user's palm is easily leaked, which poses a security risk. In the identity recognition method provided in this embodiment, Fig. 20 As shown in the figure, the user can use the palm-swipe payment method by bending the index finger to partially hide the palm, which can prevent the complete biometric features of the palm from being stolen. Fig.21 As shown, the user can also use different hand shapes, specifically the index finger, middle finger, ring finger and little finger are all bent inward to perform palm payment, thereby preventing the complete biometric features of the palm from being stolen and expanding the hand shapes for palm payment. Even if the information of a certain hand shape is leaked, the security of palm payment can be ensured by changing the hand shape. The processing of identity recognition through multiple palm hand shape patterns provided in this embodiment can enable the user to have "multiple palms", just like multiple cards or multiple passwords, making the biometric features of the user's palm more secure.

[0160] The present application also provides an application scenario, which applies the above-mentioned identity recognition method. Specifically, the application of the identity recognition method in the application scenario is as follows:

[0161] In order to solve the current situation of minors being addicted to online games, the online game anti-addiction system is currently used to limit the online game time of minors, specifically through various online game time regulations to limit the online game time of minors. For example, in the online game anti-addiction system, the cumulative game time of minors within 3 hours is "healthy" game time; the 2 hours of game time after 3 hours is "fatigue" time, during which the game income obtained by the player will be halved; if the cumulative game time exceeds 5 hours, it is "unhealthy" game time, and the player's income is reduced to 0, thereby forcing minors to go offline to rest and study, and realizing the restriction of minors' online game time. For another example, each online game can only provide 1 hour of service to minors from 20:00 to 21:00 every day on Fridays, Saturdays, Sundays and statutory holidays. At this time, in addition to determining the online game time, it is also necessary to identify the identity of the game user to determine whether he is a minor, so as to determine whether the game user's game time needs to be restricted.

[0162] For example, from 20:00 to 21:00 on Friday, if the online game time of a game user has reached 1 hour, if the game user continues to play the game, the game user's identity needs to be identified to determine whether the game user is a minor. If so, the game user is restricted and the game service is not provided to the minor. Specifically, the online game triggers an identity recognition trigger event, and the terminal responds to the identity recognition trigger event, turns on the camera, collects biometric data for the target part of the game user that is in a partially hidden and deformed form, such as the face or palm or other parts, and extracts biometric features from the collected biometric data to obtain a biometric feature combination corresponding to the game user, and the biometric feature combination includes at least two types of biometric features at the target part of the target user. The terminal performs feature matching on the collected biometric feature combination with the pre-stored registered biometric feature pattern to obtain a biometric feature matching result. The registered biometric feature pattern is a biometric feature combination obtained by identity registration of the target part of the registered user when the target part of the registered user is in a partially hidden and deformed form. The terminal determines the identification result for the game user according to the biometric matching result. When the game user is determined to be a minor according to the identification result, the online game service is stopped to the game user to prevent the minor from continuing to play online games by pretending to be an adult; if the game user is determined to be an adult according to the identification result, the online game service can be continued to be provided to the game user, thereby realizing the anti-addiction treatment of online games based on identity identification. It should be understood that although the steps in the flowcharts involved in the embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless clearly stated in this article, the execution of these steps is not strictly restricted in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the embodiments described above may include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0163] In one embodiment, Fig. 22 As shown, an identity recognition device 2200 is provided. The device can adopt a software module or a hardware module, or a combination of the two to become a part of a computer device. The device specifically includes: a biometric feature collection module 2202, a biometric feature matching module 2204 and an identity recognition processing module 2206, wherein:

[0164] The biometric feature collection module 2202 is used to collect a biometric feature combination of a target part of a target user when the target part is in a partially hidden deformed state in response to an identity recognition trigger event; the biometric feature combination is a combination of at least two types of biometric features at the target part;

[0165] The biometric matching module 2204 is used to perform feature matching between the collected biometric combination and the pre-stored registered biometric pattern to obtain a biometric matching result; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed state;

[0166] The identity recognition processing module 2206 is used to determine the identity recognition result for the target user according to the biometric matching result.

[0167] In one embodiment, the target part is the palm; the biometric feature combination includes at least the palm print feature and the palm vein feature at the palm; the biometric feature matching module 2204 includes a registration mode determination module, a palm print matching module, a palm vein matching module and a matching result acquisition module; wherein: the registration mode determination module is used to determine the pre-stored registration biometric feature mode; the registration biometric feature mode includes at least the palm print registration feature and the palm vein registration feature obtained by identity registration of the palm of the registered user; the palm print matching module is used to perform palm print feature matching on the palm print registration feature of the collected biometric feature combination to obtain a palm print feature matching result; the palm vein matching module is used to perform palm vein feature matching on the palm vein registration feature of the collected biometric feature combination to obtain a palm vein feature matching result; the matching result acquisition module is used to obtain a biometric feature matching result according to the palm print feature matching result and the palm vein feature matching result.

[0168] In one embodiment, the matching result acquisition module includes a matching weight module and a weighted fusion module; wherein: the matching weight module is used to determine the palm print matching weight and the palm vein matching weight; the weighted fusion module is used to perform weighted fusion on the palm print feature matching result and the palm vein feature matching result based on the palm print matching weight and the palm vein matching weight to obtain the biometric matching result.

[0169] In one embodiment, the target part is the palm; the biometric feature acquisition module 2202 includes an identification trigger response module, a feature map acquisition module and a feature extraction module; wherein: the identification trigger response module is used to enter a palm feature acquisition state in response to an identity recognition trigger event; the feature map acquisition module is used to collect a biometric feature map of the target user's palm in the palm feature acquisition state; the feature extraction module is used to extract a biometric feature combination of the target user's palm from the biometric map when it is determined that the target user's palm is in a partially hidden palm shape.

[0170] In one embodiment, the feature extraction module includes a morphology recognition module, a palm feature extraction module and a palm feature combination acquisition module; wherein: the morphology recognition module is used to perform palm morphology recognition on the palm of the target user based on the biometric map to obtain a palm morphology recognition result; the palm feature extraction module is used to extract at least two types of palm biometric features of the palm of the target user from the biometric map when the palm morphology recognition result indicates that the palm of the target user is in a partially hidden palm shape; the palm feature combination acquisition module is used to obtain a biometric feature combination of the palm of the target user based on at least two types of palm biometric features.

[0171] In one embodiment, it also includes a gesture determination module and a gesture recognition module; wherein: the gesture determination module is used to determine the target gesture corresponding to the palm of the target user according to the biometric map; the gesture recognition module is used to perform gesture recognition on the target gesture based on each registered gesture in the local registered gesture library to obtain a gesture recognition result; the biometric matching module 2204 is also used to send the collected biometric combination to the server when the gesture recognition result is recognition passed, so that the server can perform feature matching on the collected biometric combination with the pre-stored registered biometric pattern; and obtain the biometric matching result returned by the server.

[0172] In one embodiment, it also includes a registration event response module, a registration feature acquisition module and a registration processing module; wherein: the registration event response module is used to enter a biometric feature collection state in response to a triggering event of registration identity recognition; the registration feature acquisition module is used to collect a biometric feature combination to be registered when the target part of the registered user is in a partially hidden deformed form in the biometric feature collection state; the registration processing module is used to perform identity registration on the target part of the registered user based on the biometric feature combination to be registered when the biometric feature combination to be registered meets the identity recognition registration conditions, and obtain a registered biometric feature pattern.

[0173] In one embodiment, it also includes a replacement event response module, a replacement feature acquisition module and a replacement processing module; wherein: the replacement event response module is used to respond to the triggering event of replacing the registered biometric feature pattern, and collect the biometric feature combination to be replaced when the target part of the registered user is in a partially hidden deformed form; the replacement feature acquisition module is used to perform feature validity analysis on various types of biometric features in the biometric feature combination to be replaced to obtain a feature analysis result; the replacement processing module is used to replace the biometric feature combination in the registered biometric feature pattern with the biometric feature combination to be replaced when the feature analysis result indicates that the biometric feature combination to be replaced is valid.

[0174] In one embodiment, it also includes a resource transfer parameter determination module, a resource account determination module and a resource transfer processing module; wherein: the resource transfer parameter determination module is used to determine the resource transfer parameters in response to a resource transfer triggering event; the resource account determination module is used to determine the target resource account associated with the target user according to the identity recognition result; the resource transfer processing module is used to transfer resources to the target resource account based on the resource transfer parameters.

[0175] In one embodiment, the biometric matching module 2204 includes a feature pattern determination module, a feature pattern matching module and a feature matching result acquisition module; wherein: the feature pattern determination module is used to determine the pre-stored registered biometric pattern; the feature pattern matching module is used to feature match each type of biometric in the collected biometric combination with the registered biometric of the corresponding type of each type of biometric in the registered biometric pattern to obtain the feature matching results corresponding to each type of biometric; the feature matching result acquisition module is used to obtain the biometric matching result according to the feature matching results corresponding to each type of biometric.

[0176] In one embodiment, the biometric matching module 2204 includes a scene determination module, a scene feature pattern determination module and a scene feature pattern matching module; wherein: the scene determination module is used to determine the trigger scene corresponding to the identity recognition trigger event; the scene feature pattern determination module is used to determine the target registered biometric pattern associated with the trigger scene from the pre-stored registered biometric patterns; the scene feature pattern matching module is used to combine the collected biometrics and perform feature matching with the target registered biometric pattern to obtain a biometric matching result.

[0177] In one embodiment, the registered biometric pattern includes at least two biometric combinations obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in different deformed forms of local concealment; the biometric matching module 2204 is also used to perform feature matching on the collected biometric combination with various biometric combinations in the pre-stored registered biometric pattern, and obtain the biometric matching results corresponding to each biometric combination in the registered biometric pattern; the identity recognition processing module 2206 is also used to determine the identity recognition result for the target user based on the successfully matched biometric matching results among the biometric matching results corresponding to the various biometric combinations.

[0178] In one embodiment, it also includes a feature image acquisition module, a local hidden analysis module and an identification failure determination module; wherein: the feature image acquisition module is used to collect a feature image of a target part of a target user in response to an identity recognition trigger event; the local hidden analysis module is used to perform a local hidden analysis on the target part based on the feature image to obtain a local hidden analysis result; the identification failure determination module is used to determine that the identity recognition result of the target user is an identity recognition failure when the local hidden analysis result indicates that the target part is not in a locally hidden deformed form.

[0179] In one embodiment, it also includes a depth image acquisition module, a liveness detection module and a detection result processing module; wherein: the depth image acquisition module is used to collect a depth image corresponding to a target part of a target user in response to an identity recognition trigger event; the liveness detection module is used to perform liveness detection on the target user based on the depth image to obtain a liveness detection result; the detection result processing module is used to determine that the identity recognition result of the target user is an identity recognition failure when the liveness detection result is a detection failure.

[0180] For the specific definition of the identity recognition device, please refer to the definition of the identity recognition method above, which will not be repeated here. Each module in the above-mentioned identity recognition device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0181] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Fig.23As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The computer device also includes an input and output interface, which is a connection circuit for exchanging information between the processor and the external device. They are connected to the processor through a system bus, referred to as an I / O interface. When the computer program is executed by the processor, an identity recognition method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0182] Those skilled in the art will understand that Fig.23 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0183] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0184] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0185] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0186] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0187] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0188] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0189] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. An identity recognition method, characterized in that: The method comprises: In response to an identity recognition triggering event, for a target part of a target user, a biometric feature combination is collected when the target part is in a partially hidden deformed state; the biometric feature combination is a combination of at least two types of biometric features at the target part; when the target part is in a partially hidden deformed state, the global biometric feature of the target part is partially hidden; The collected biometric combination is matched with various biometric combinations in a pre-stored registered biometric pattern to obtain biometric matching results corresponding to each biometric combination in the registered biometric pattern; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form; the registered biometric pattern includes at least two biometric combinations obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in different partially hidden deformed forms; The identity recognition result for the target user is determined based on the successfully matched biometric matching results among the biometric matching results corresponding to the various biometric combinations.

2. The method according to claim 1, characterized in that The target part is a palm; the biometric feature combination includes at least a palm print feature and a palm vein feature at the palm; the collected biometric feature combination is matched with various biometric feature combinations in a pre-stored registered biometric feature pattern to obtain biometric feature matching results corresponding to each biometric feature combination in the registered biometric feature pattern, including: Determine a pre-stored registration biometric feature pattern; the registration biometric feature pattern at least includes a palm print registration feature and a palm vein registration feature obtained by performing identity registration on the palm of the registered user; Performing palmprint feature matching on the palmprint feature in the collected biometric feature combination and the palmprint registration feature to obtain a palmprint feature matching result; Performing palm vein feature matching on the palm vein feature in the collected biometric feature combination and the palm vein registration feature to obtain a palm vein feature matching result; A biometric feature matching result is obtained according to the palm print feature matching result and the palm vein feature matching result.

3. The method according to claim 2, characterized in that The obtaining of a biometric feature matching result according to the palm print feature matching result and the palm vein feature matching result includes: Determine palm print matching weight and palm vein matching weight; Based on the palm print matching weight and the palm vein matching weight, the palm print feature matching result and the palm vein feature matching result are weightedly fused to obtain a biometric feature matching result.

4. The method according to claim 1, characterized in that: The target part is a palm; in response to the identity recognition triggering event, for the target part of the target user, collecting a combination of biometric features when the target part is in a partially hidden deformed state, including: In response to an identity recognition trigger event, entering a palm feature collection state; In the palm feature collection state, collecting a biometric feature map of the palm of the target user; When it is determined that the palm of the target user is in a partially hidden palm state, a biometric feature combination of the palm of the target user is extracted from the biometric feature map.

5. The method according to claim 4, characterized in that When it is determined that the palm of the target user is in a partially hidden palm state, extracting a biometric combination of the palm of the target user from the biometric map includes: Performing palm shape recognition on the palm of the target user based on the biometric feature image to obtain a palm shape recognition result; When the palm shape recognition result indicates that the palm of the target user is in a partially hidden palm shape, extracting at least two types of palm biometric features of the palm of the target user from the biometric feature map; A combination of palm biometric features of the target user is obtained according to the at least two types of palm biometric features.

6. The method according to claim 4, characterized in that The method further comprises: Determining a target gesture corresponding to the palm of the target user according to the biometric feature map; Based on each registered gesture in the local registered gesture library, gesture recognition is performed on the target gesture to obtain a gesture recognition result; The collected biometric combination is matched with various biometric combinations in the pre-stored registered biometric pattern to obtain biometric matching results corresponding to each biometric combination in the registered biometric pattern, including: When the gesture recognition result is recognition passed, the collected biometric feature combination is sent to the server, so that the server performs feature matching on the collected biometric feature combination and the pre-stored registered biometric feature pattern; Obtaining the biometric feature matching result returned by the server.

7. The method according to claim 1, characterized in that The method further comprises: In response to a triggering event of registration identification, entering a biometric feature collection state; In the biometric feature collection state, collecting a biometric feature combination to be registered when the target part of the registered user is in a partially hidden deformed state; When the validity score of the biometric combination to be registered is greater than the validity score threshold in the identity recognition registration condition, identity registration is performed on the target part of the registered user based on the biometric combination to be registered to obtain the registered biometric pattern.

8. The method according to claim 1, characterized in that The method further comprises: In response to a triggering event for replacing a registered biometric feature pattern, collecting a biometric feature combination to be replaced when the target part of the registered user is in a partially hidden deformed state; Performing feature validity analysis on various types of biometric features in the biometric feature combination to be replaced to obtain feature analysis results; When the feature analysis result indicates that the biometric feature combination to be replaced is valid, the biometric feature combination in the registered biometric feature pattern is replaced with the biometric feature combination to be replaced.

9. The method according to claim 1, characterized in that: The method further comprises: In response to a resource transfer triggering event, determining a resource transfer parameter; Determine the target resource account associated with the target user according to the identity recognition result; The resource transfer is performed on the target resource account based on the resource transfer parameter.

10. The method according to any one of claims 1 to 9, characterized in that: The collected biometric combination is matched with various biometric combinations in the pre-stored registered biometric pattern to obtain biometric matching results corresponding to each biometric combination in the registered biometric pattern, including: determining a pre-stored enrolled biometric pattern; Perform feature matching on each type of biometric feature in the collected biometric feature combination and the registered biometric feature of the corresponding type of each type of biometric feature in the registered biometric feature pattern to obtain feature matching results corresponding to each type of biometric feature; A biometric feature matching result is obtained according to the feature matching results corresponding to each type of the biometric feature.

11. The method according to any one of claims 1 to 9, characterized in that: The collected biometric combination is matched with various biometric combinations in the pre-stored registered biometric pattern to obtain biometric matching results corresponding to each biometric combination in the registered biometric pattern, including: Determine a triggering scenario corresponding to the identity recognition triggering event; Determining a target registered biometric pattern associated with the triggering scenario from pre-stored registered biometric patterns; The collected biometric combination is matched with the target registered biometric pattern to obtain a biometric matching result.

12. The method according to any one of claims 1 to 9, characterized in that: The method further comprises: In response to an identity recognition triggering event, collecting a feature image of the target part of the target user; Performing local hidden analysis on the target part based on the characteristic image to obtain a local hidden analysis result; When the local hiding analysis result indicates that the target part is not in a deformed state of local hiding, it is determined that the identity recognition result of the target user is an identity recognition failure.

13. The method according to any one of claims 1 to 9, characterized in that: The method further comprises: In response to an identity recognition trigger event, acquiring a depth image corresponding to a target part of the target user; Performing liveness detection on the target user based on the depth image to obtain a liveness detection result; When the liveness detection result is detection failure, it is determined that the identity recognition result of the target user is identity recognition failure.

14. An identity recognition device, characterized in that: The device comprises: A biometric feature collection module, configured to collect, in response to an identity recognition trigger event, a combination of biometric features of a target part of a target user when the target part is in a partially hidden deformed state; the combination of biometric features is a combination of at least two types of biometric features at the target part; when the target part is in a partially hidden deformed state, the global biometric features of the target part are partially hidden; A biometric matching module, used to perform feature matching on the collected biometric combination with various biometric combinations in a pre-stored registered biometric pattern, and obtain biometric matching results corresponding to each biometric combination in the registered biometric pattern; the registered biometric pattern is a biometric combination obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in a partially hidden deformed form; the registered biometric pattern includes at least two biometric combinations obtained by performing identity registration on the target part of the registered user when the target part of the registered user is in different partially hidden deformed forms; The identity recognition processing module is used to determine the identity recognition result for the target user according to the successfully matched biometric matching results among the biometric matching results corresponding to the various biometric combinations.

15. The device according to claim 14, characterized in that The target part is a palm; the biometric feature combination includes at least a palm print feature and a palm vein feature at the palm; The biometric matching module is further used to determine a pre-stored registered biometric pattern; the registered biometric pattern at least includes a palm print registration feature and a palm vein registration feature obtained by performing identity registration on the palm of a registered user; the palm print feature in the collected biometric combination is matched with the palm print registration feature to obtain a palm print feature matching result; the palm vein feature in the collected biometric combination is matched with the palm vein registration feature to obtain a palm vein feature matching result; and a biometric matching result is obtained based on the palm print feature matching result and the palm vein feature matching result.

16. The device according to claim 15, characterized in that The biometric matching module is further used to determine a palm print matching weight and a palm vein matching weight; based on the palm print matching weight and the palm vein matching weight, the palm print feature matching result and the palm vein feature matching result are weightedly fused to obtain a biometric matching result.

17. The device according to claim 14, characterized in that The target part is the palm; The biometric feature collection module is also used to enter a palm feature collection state in response to an identity recognition trigger event; in the palm feature collection state, collect a biometric feature map of the target user's palm; when it is determined that the palm of the target user is in a partially hidden palm shape, extract the biometric feature combination of the target user's palm from the biometric feature map.

18. The device according to claim 17, characterized in that The biometric feature acquisition module is further used to perform palm shape recognition on the palm of the target user based on the biometric feature graph to obtain a palm shape recognition result; when the palm shape recognition result indicates that the palm of the target user is in a partially hidden palm shape, extract at least two types of palm biometric features of the palm of the target user from the biometric feature graph; A combination of palm biometric features of the target user is obtained according to the at least two types of palm biometric features.

19. The device according to claim 17, characterized in that The device also includes: A gesture determination module, used to determine a target gesture corresponding to the palm of the target user according to the biometric feature map; A gesture recognition module, used to perform gesture recognition on the target gesture based on each registered gesture in a local registered gesture library to obtain a gesture recognition result; The biometric matching module is also used to send the collected biometric combination to the server when the gesture recognition result is recognition passed, so that the server can perform feature matching on the collected biometric combination with the pre-stored registered biometric pattern; and obtain the biometric matching result returned by the server.

20. The device according to claim 14, characterized in that The device also includes: A registration event response module, used to respond to a triggering event of registration identification and enter a biometric feature collection state; A registration feature acquisition module, used for collecting a combination of biometric features to be registered when the target part of the registered user is in a partially hidden deformed state in the biometric feature collection state; The registration processing module is used to perform identity registration on the target part of the registered user based on the biometric combination to be registered when the validity score of the biometric combination to be registered is greater than the validity score threshold in the identity recognition registration condition, so as to obtain the registered biometric pattern.

21. The device according to claim 14, characterized in that The device also includes: a replacement event response module, configured to respond to a triggering event of replacing a registered biometric feature pattern and collect a combination of biometric features to be replaced when the target part of the registered user is in a partially hidden deformed state; A replacement feature acquisition module, used to perform feature validity analysis on various types of biometric features in the biometric feature combination to be replaced, and obtain feature analysis results; The replacement processing module is used to replace the biometric feature combination in the registered biometric feature pattern with the biometric feature combination to be replaced when the feature analysis result indicates that the biometric feature combination to be replaced is valid.

22. The device according to claim 14, characterized in that The device also includes: A resource transfer parameter determination module, configured to determine a resource transfer parameter in response to a resource transfer triggering event; A resource account determination module, used to determine a target resource account associated with the target user according to the identity recognition result; The resource transfer processing module is used to transfer resources to the target resource account based on the resource transfer parameters.

23. The device according to any one of claims 14 to 22, characterized in that The biometric matching module is also used to determine a pre-stored registered biometric pattern; perform feature matching on each type of biometric in the collected biometric combination with the registered biometric of the corresponding type of each type of biometric in the registered biometric pattern to obtain feature matching results corresponding to each type of biometric; and obtain a biometric matching result based on the feature matching results corresponding to each type of biometric.

24. The device according to any one of claims 14 to 22, characterized in that The biometric matching module is also used to determine the triggering scenario corresponding to the identity recognition triggering event; determine the target registered biometric pattern associated with the triggering scenario from the pre-stored registered biometric patterns; and perform feature matching on the collected biometric combination and the target registered biometric pattern to obtain a biometric matching result.

25. The device according to any one of claims 14 to 22, characterized in that The device also includes: A feature image acquisition module, configured to acquire a feature image of the target part of the target user in response to an identity recognition trigger event; A local hidden analysis module, used to perform local hidden analysis on the target part based on the feature image to obtain a local hidden analysis result; The recognition failure determination module is used to determine that the identity recognition result of the target user is an identity recognition failure when the local hiding analysis result indicates that the target part is not in a deformed state of local hiding.

26. The device according to any one of claims 14 to 22, characterized in that The device also includes: A depth image acquisition module, configured to acquire a depth image corresponding to a target part of the target user in response to an identity recognition trigger event; A liveness detection module, used to perform liveness detection on the target user based on the depth image to obtain a liveness detection result; The detection result processing module is used to determine that the identity recognition result of the target user is an identity recognition failure when the liveness detection result is a detection failure.

27. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 13 are implemented.

28. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

29. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

Images