Biometric identification method and device

Abstract

The embodiment of the present application provides a biometric identification method and device, which can be used in the field of artificial intelligence technology. The method includes: inputting the environmental feature data corresponding to the target user who initiates the biometric identification request into a preset multi-algorithm decision model, and selecting one of multiple preset biometric identification algorithms as the target biometric identification algorithm corresponding to the target user according to the output of the multi-algorithm decision model; and obtaining the target biometric identification result corresponding to the biological image data of the target user based on the target biometric identification algorithm. The present application can effectively improve the degree of automation and intelligence in selecting the target biometric identification algorithm in the multi-biometric identification algorithm, and thus can effectively improve the reliability and accuracy of biometric identification based on the target biometric identification algorithm, and effectively improve the efficiency of biometric identification.

Description

Technical Field

[0001] The present application relates to the field of biometric identification technology, in particular to the field of artificial intelligence technology, and specifically to a method and device for biometric feature identification. Background Art

[0002] Biometric technology has been widely used in commercial banks' product services, customer marketing, operations management, and risk control, such as face-swiping payment, counter identity verification, fingerprint / biometric access control / attendance, and other scenarios.

[0003] At present, traditional biometric systems use a single algorithm model, but are applied in various scenarios at the same time, which makes the model performance better in a single scenario, but the performance drops sharply in other complex scenarios. Therefore, major financial institutions are no longer limited to using a single manufacturer's biometric algorithm, but use a new model that integrates multiple algorithms. However, different versions of biometric algorithms from different manufacturers have their own advantages and disadvantages in specific scenarios. Therefore, the current selection of multiple biometric algorithms is based on manual adjustment of expert rules, and it is impossible to integrate scene characteristics to automatically select the optimal algorithm for application. Summary of the invention

[0004] In response to the problems in the prior art, the present application provides a biometric recognition method and device, which can effectively improve the degree of automation and intelligence in selecting a target biometric recognition algorithm among multiple biometric recognition algorithms, thereby effectively improving the reliability and accuracy of biometric recognition based on the target biometric recognition algorithm, and effectively improving the efficiency of biometric recognition.

[0005] In order to solve the above technical problems, this application provides the following technical solutions:

[0006] In a first aspect, the present application provides a biometric recognition method, comprising:

[0007] Inputting the environmental feature data corresponding to the target user who initiates the biometric feature recognition request into a preset multi-algorithm decision model, and selecting one of the multiple preset biometric feature recognition algorithms as the target biometric feature recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model;

[0008] A target biometric feature recognition result corresponding to the biometric image data of the target user is obtained based on the target biometric feature recognition algorithm.

[0009] Furthermore, the environmental characteristic data includes: customer profile data, business requirement data and algorithm capability data;

[0010] Correspondingly, before inputting the environmental feature data corresponding to the target user initiating the biometric feature recognition request into the preset multi-algorithm decision model, it also includes:

[0011] Receiving a biometric identification request and corresponding biometric data of a target user, wherein the biometric identification request includes a unique identifier of the target user;

[0012] Based on the unique identifier of the target user, the customer portrait data corresponding to the target user is obtained, and the pre-stored business requirement data and algorithm capability data are retrieved.

[0013] Furthermore, the step of inputting the environmental feature data corresponding to the target user initiating the biometric feature recognition request into a preset multi-algorithm decision model, and selecting one of a plurality of preset biometric feature recognition algorithms as the target biometric feature recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model, includes:

[0014] Acquire the target algorithm identifier output by the multi-algorithm decision model according to the customer profile data, business requirement data and algorithm capability data corresponding to the target user;

[0015] The biometric recognition algorithm corresponding to the target algorithm identifier is determined from the one-to-one correspondence between each pre-stored algorithm identifier and each biometric recognition algorithm, and the biometric recognition algorithm is determined as the target biometric recognition algorithm corresponding to the target user.

[0016] Furthermore, the environmental feature data also includes: weights corresponding to the customer profile data, the business requirement data, and the algorithm capability data;

[0017] Correspondingly, the step of obtaining the target algorithm identifier output by the multi-algorithm decision model according to the customer profile data, business requirement data, and algorithm capability data corresponding to the target user includes:

[0018] The customer portrait data, business requirement data and algorithm capability data corresponding to the target user, as well as the weights corresponding to the customer portrait data, business requirement data and algorithm capability data corresponding to the target user are input into the multi-algorithm decision model so that the multi-algorithm decision model outputs a target algorithm identifier.

[0019] Furthermore, it also includes:

[0020] Obtaining a training data set, wherein the training data set includes historical environmental feature data of multiple users and labels corresponding to the historical environmental feature data of each user, wherein the label is one of algorithm identifiers corresponding to each preset biometric feature recognition algorithm;

[0021] The training data set is trained with a deep learning network algorithm to obtain a multi-algorithm decision model for selecting each of the biometric feature recognition algorithms and outputting a corresponding algorithm identifier.

[0022] Furthermore, the historical environment characteristic data includes: customer portrait data, business requirement data and algorithm capability data, and the historical environment characteristic data also includes: weights corresponding to the customer portrait data, business requirement data and algorithm capability data respectively.

[0023] Furthermore, after receiving the biometric feature recognition request and the corresponding biometric collection data of the target user, the method further includes:

[0024] If the target user's biological data is image data, it is determined whether the image data meets the preset biometric image quality requirements. If so, the target user's biological data is preprocessed to obtain the target user's biological image data.

[0025] Furthermore, after receiving the biometric feature recognition request and the corresponding biometric collection data of the target user, the method further includes:

[0026] If the biological data collected by the target user is video data, performing frame extraction processing on the video data based on a preset frame extraction rule to obtain a plurality of image data corresponding to the target user;

[0027] It is determined whether there is image data that meets the preset biometric image quality requirements among the image data. If so, the image data that meets the biometric image quality requirements are preprocessed to obtain the biometric image data of the target user.

[0028] Furthermore, it also includes:

[0029] Obtaining a biometric processing type corresponding to the biometric identification request;

[0030] The target biometric feature recognition result corresponding to the biometric image data of the target user is processed based on the biometric feature processing type, and the corresponding processing result is output.

[0031] Further, the biometric processing types include: biometric registration;

[0032] Correspondingly, the target biometric feature recognition result corresponding to the biometric image data of the target user is processed based on the biometric feature processing type, including:

[0033] The target biometric feature recognition result corresponding to the biometric image data of the target user is stored in at least one biometric feature recognition result corresponding to the identifier of the target user to complete the biometric feature registration for the target biometric feature recognition result.

[0034] Further, the biometric processing types include: one-to-one identification;

[0035] A target biometric feature recognition result corresponding to the biometric image data of the target user is identified one-to-one with a pre-stored biometric feature recognition result corresponding to the identifier of the target user to obtain a corresponding one-to-one recognition result.

[0036] Further, the biometric processing types include: one-to-many identification;

[0037] A target biometric feature recognition result corresponding to the biometric image data of the target user is subjected to one-to-many recognition with a plurality of pre-stored biometric feature recognition results corresponding to the identifier of the target user to obtain a corresponding one-to-many recognition result.

[0038] In a second aspect, the present application provides a biometric feature recognition device, comprising:

[0039] A decision module, used to input the environmental feature data corresponding to the target user who initiates the biometric feature recognition request into a preset multi-algorithm decision model, and select one of the multiple preset biometric feature recognition algorithms as the target biometric feature recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model;

[0040] The recognition module is used to obtain the target biometric feature recognition result corresponding to the biological image data of the target user based on the target biometric feature recognition algorithm.

[0041] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the biometric recognition method when executing the program.

[0042] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the biometric recognition method when executed by a processor.

[0043] It can be seen from the above technical scheme that the present application provides a biometric identification method and device, the method comprising: inputting the environmental feature data corresponding to the target user who initiates the biometric identification request into a preset multi-algorithm decision model, and selecting one of multiple preset biometric identification algorithms as the target biometric identification algorithm corresponding to the target user according to the output of the multi-algorithm decision model; obtaining the target biometric identification result corresponding to the biological image data of the target user based on the target biometric identification algorithm; inputting the environmental feature data corresponding to the target user who initiates the biometric identification request into the preset multi-algorithm decision model, and selecting one of multiple preset biometric identification algorithms as the target biometric identification algorithm corresponding to the target user according to the output of the multi-algorithm decision model. It can effectively realize dynamic decision-making of multiple biometric algorithms that integrate scene characteristics without human intervention, and can effectively improve the degree of automation and intelligence in selecting target biometric recognition algorithms among multiple biometric recognition algorithms, thereby effectively improving the reliability and accuracy of biometric recognition based on the target biometric recognition algorithm, and effectively improving the efficiency of biometric recognition; at the same time, it makes up for the defect that traditional biometric systems can only use a single algorithm for recognition, and provides an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the system's generalization ability; it can effectively reduce the cost of manual adjustment when applying multiple algorithms in emerging biometric systems, greatly reduce the risks of easy errors and impact on recognition pass rate caused by manual adjustments, and improve the intelligence level of biometric systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 It is a first flow chart of the biometric identification method in the embodiment of the present application.

[0046] Figure 2 It is a second flow chart of the biometric identification method in the embodiment of the present application.

[0047] Figure 3 This is a third flow chart of the biometric identification method in the embodiment of the present application.

[0048] Figure 4 This is a fourth flow chart of the biometric identification method in the embodiment of the present application.

[0049] Figure 5This is a fifth flow chart of the biometric identification method in the embodiment of the present application.

[0050] Figure 6 This is a sixth flow chart of the biometric feature recognition method in the embodiment of the present application.

[0051] Figure 7 This is the seventh flow chart of the biometric identification method in the embodiment of the present application.

[0052] Figure 8 This is the eighth flow chart of the biometric identification method in the embodiment of the present application.

[0053] Fig. 9 This is the ninth flow chart of the biometric identification method in the embodiment of the present application.

[0054] Fig.10 This is the tenth flow chart of the biometric identification method in the embodiment of the present application.

[0055] Fig.11 This is the eleventh flow chart of the biometric identification method in the embodiment of the present application.

[0056] Fig.12 It is a schematic diagram of the structure of the biometric identification device in the embodiment of the present application.

[0057] Fig.13 This is a schematic diagram of the structure of the biometric recognition system with multi-algorithm intelligent decision-making in this application example.

[0058] Fig.14 This is a schematic diagram of the structure of the biometric feature acquisition module in the multi-algorithm intelligent decision-making biometric recognition system in this application example.

[0059] Fig.15 This is a schematic diagram of the structure of the data transmission module in the biometric recognition system with multi-algorithm intelligent decision-making in this application example.

[0060] Fig.16 This is a structural diagram of the biometric multi-algorithm intelligent decision-making module in the biometric system with multi-algorithm intelligent decision-making in this application example.

[0061] Fig.17 The execution flow chart of the biometric main control module provided for this application example.

[0062] Fig.18 The upload execution flow chart of the data transmission module provided for this application example.

[0063] Fig.19 This is the execution flow chart of the data transmission module provided for this application example.

[0064] Fig. 20 Execution flow chart of the biometric multi-algorithm intelligent decision-making module provided for this application example.

[0065] Fig.21 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0067] It should be noted that the biometric identification method and device disclosed in this application can be used in the field of artificial intelligence technology, and can also be used in any field outside the field of artificial intelligence technology. The application field of the biometric identification method and device disclosed in this application is not limited.

[0068] In view of the problems that the existing selection of multiple biometric recognition algorithms relies on manual labor, has poor reliability and low efficiency, the present application provides a biometric recognition method, which inputs the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and selects one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model; obtains the target biometric recognition result corresponding to the biological image data of the target user based on the target biometric recognition algorithm, and inputs the environmental feature data corresponding to the target user who initiates the biometric recognition request into the preset multi-algorithm decision model, and selects one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model. The method can effectively realize the dynamic decision of biometric multi-algorithms integrating scene characteristics without manual participation, making up for the defect that the traditional biometric recognition system can only use a single algorithm for recognition, and provides an intelligent decision-making solution for the simultaneous decision-making of multiple algorithms, greatly improving the generalization ability of the system; it can effectively reduce the cost of manual adjustment when applying multiple algorithms in emerging biometric recognition systems, greatly reduce the risks of easy errors and impact on recognition pass rate caused by manual adjustment, and improve the intelligence level of the biometric recognition system.

[0069] Based on the above content, the present application also provides a biometric recognition device for implementing the biometric recognition method provided in one or more embodiments of the present application. The biometric recognition device can communicate and connect with the client devices held by technicians and users by itself or through a third-party server, etc. The biometric recognition device can receive biometric recognition requests or instructions sent by technicians and users through client devices held by them, and then input the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and select one from multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user based on the output of the multi-algorithm decision model; based on the target biometric recognition algorithm, obtain the target biometric recognition result corresponding to the biological image data of the target user, and send the target biometric recognition result to the technician or user through the client device held, and can also be sent to other service systems of enterprises such as financial institutions for subsequent processing.

[0070] In one or more embodiments of the present application, the biometric recognition device may be a functional module provided in a service system of an enterprise such as a financial institution, or the biometric recognition device may be separately deployed as a server or other device that may interact with the service system.

[0071] The biometric identification part of the aforementioned biometric identification device can be executed in the server as described above. In another practical application scenario, all operations can also be completed in the client device. The specific selection can be based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor for specific processing of biometric identification.

[0072] It is understandable that the client device used to send a request or instruction to the biometric recognition device can be a device with an image acquisition function, or can be connected to other image acquisition devices to acquire biological images or videos of the user. The client device can include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device can include smart glasses, smart watches, smart bracelets, etc.

[0073] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0074] The server and the client device may communicate with each other using any suitable network protocol, including network protocols that have not yet been developed on the date of filing this application. The network protocols may include, for example, TCP / IP, UDP / IP, HTTP, HTTPS, etc. Of course, the network protocols may also include, for example, RPC (Remote Procedure Call Protocol) and REST (Representational State Transfer) protocols used on top of the above protocols.

[0075] The details are described in detail through the following embodiments and application examples.

[0076] In order to solve the problems of existing multi-biometric recognition algorithm selection relying on manual work, poor reliability and low efficiency, the present application provides an embodiment of a biometric recognition method that can be performed by a biometric recognition device, see Figure 1 The biometric feature recognition method specifically includes the following contents:

[0077] Step 100: Input the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and select one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user based on the output of the multi-algorithm decision model.

[0078] In step 100, a biometric identification request may be sent by a technical staff of a financial institution or a financial user through a client device held by the user to a biometric identification device. It will be understood that the biometrics mentioned in one or more embodiments of the present application refer to physiological characteristics (fingerprints, irises, facial features, etc.) or behavioral characteristics (gait, etc.) inherent in the human body.

[0079] Step 200: Obtain a target biometric feature recognition result corresponding to the biometric image data of the target user based on the target biometric feature recognition algorithm.

[0080] It can be understood that the multi-algorithm decision mentioned in one or more embodiments of the present application may also refer to a multi-algorithm decision mechanism based on expert rules or a multi-algorithm decision mechanism based on voting, but in a preferred embodiment of the present application, the execution method of step 100 is preferentially selected to effectively improve the degree of automation and intelligence of multi-algorithm decision and biometric recognition.

[0081] In step 200, biometric features refer in particular to feature data that can be collected through images or videos. For example, the biometric image data can be facial image data, iris image data, retinal image data, facial image data, vein distribution map, palm three-dimensional image data, and signature image data of the target user; if the collected data is video data, the target user corresponds to multiple biometric image data obtained by extracting key frames from the video data.

[0082] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application, by inputting the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and selecting one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model, can effectively realize the dynamic decision-making of biometric multi-algorithms integrating scene characteristics without human participation, make up for the defect that traditional biometric systems can only use a single algorithm for recognition, and provide an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the generalization ability of the system; it can effectively reduce the cost of manual adjustment when applying multiple algorithms in emerging biometric systems, greatly reduce the risks of easy errors and impact on recognition pass rate caused by manual adjustments, and improve the intelligence level of biometric systems.

[0083] In order to effectively improve the efficiency and comprehensiveness of obtaining environmental feature data, in one embodiment of the biometric feature recognition method provided in this application, the environmental feature data includes: customer portrait data, business requirement data and algorithm capability data; see Figure 2 The biometric feature recognition method may further include the following contents before step 100:

[0084] Step 010: Receive a biometric identification request and corresponding biometric data of a target user, wherein the biometric identification request includes a unique identifier of the target user.

[0085] Step 020: Based on the unique identifier of the target user, obtain the customer portrait data corresponding to the target user, and retrieve the pre-stored business requirement data and algorithm capability data.

[0086] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can effectively improve the efficiency and comprehensiveness of obtaining environmental feature data by obtaining the customer portrait data corresponding to the target user based on the unique identifier of the target user, and retrieving pre-stored business requirement data and algorithm capability data, thereby effectively improving the efficiency and accuracy of subsequent application of environmental feature data for multi-algorithm decision-making.

[0087] In order to effectively improve the efficiency and accuracy of the application of the multi-algorithm decision model, in one embodiment of the biometric recognition method provided in this application, see Figure 3 , step 100 of the biometric feature recognition method specifically includes the following contents:

[0088] Step 110: Obtain the target algorithm identifier output by the multi-algorithm decision model based on the customer portrait data, business requirement data and algorithm capability data corresponding to the target user.

[0089] Step 120: Determine the biometric recognition algorithm corresponding to the target algorithm identifier from the one-to-one correspondence between each pre-stored algorithm identifier and each biometric recognition algorithm, and determine the biometric recognition algorithm as the target biometric recognition algorithm corresponding to the target user.

[0090] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can effectively improve the efficiency and accuracy of the application of the multi-algorithm decision model by obtaining the target algorithm identifier output by the multi-algorithm decision model based on the customer portrait data, business requirement data and algorithm capability data corresponding to the target user, thereby effectively improving the efficiency and accuracy of biometric recognition based on the target biometric recognition algorithm.

[0091] In order to further improve the reliability and accuracy of the application of the multi-algorithm decision model, in one embodiment of the biometric identification method provided in the present application, the environmental feature data also includes: the weights corresponding to the customer portrait data, the business requirement data and the algorithm capability data; see Figure 4 , step 110 of the biometric feature recognition method specifically includes the following contents:

[0092] Step 111: Input the customer profile data, business requirement data and algorithm capability data corresponding to the target user, as well as the weights corresponding to the customer profile data, business requirement data and algorithm capability data corresponding to the target user, into the multi-algorithm decision model, so that the multi-algorithm decision model outputs a target algorithm identifier.

[0093] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can further improve the reliability and accuracy of the application of the multi-algorithm decision model by inputting the customer portrait data, business requirement data and algorithm capability data corresponding to the target user, as well as the weights corresponding to the customer portrait data, business requirement data and algorithm capability data corresponding to the target user into the multi-algorithm decision model, thereby further improving the reliability and accuracy of biometric recognition based on the target biometric recognition algorithm.

[0094] In order to further improve the reliability and accuracy of the application of the multi-algorithm decision model, in one embodiment of the biometric recognition method provided in this application, see Figure 5 The biometric feature recognition method may further include the following contents before step 100 or step 010:

[0095] Step 001: Obtain a training data set, wherein the training data set includes historical environmental feature data of multiple users and labels corresponding to the historical environmental feature data of each user, wherein the label is one of the algorithm identifiers corresponding to each preset biometric recognition algorithm.

[0096] Step 002: Train the training data set using a deep learning network algorithm to obtain a multi-algorithm decision model for selecting each of the biometric recognition algorithms and outputting a corresponding algorithm identifier.

[0097] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can effectively improve the application reliability and accuracy of the multi-algorithm decision model by training the training data set with a deep learning network algorithm, and thus can further improve the reliability and accuracy of biometric recognition according to the target biometric recognition algorithm.

[0098] In order to further improve the application reliability and accuracy of the multi-algorithm decision-making model obtained by training, in one embodiment of the biometric recognition method provided in the present application, the historical environmental feature data includes: customer portrait data, business requirement data and algorithm capability data, and the historical environmental feature data also includes: the weights corresponding to the customer portrait data, business requirement data and algorithm capability data.

[0099] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application, by setting the historical environmental feature data to include: customer portrait data, business requirement data and algorithm capability data, the historical environmental feature data also includes: the weights corresponding to the customer portrait data, business requirement data and algorithm capability data, can further train the application reliability and accuracy of the multi-algorithm decision model, thereby further improving the reliability and accuracy of biometric recognition according to the target biometric recognition algorithm.

[0100] In order to effectively improve the application of data collection, in one embodiment of the biometric recognition method provided in this application, see Figure 6 The biometric feature recognition method further includes the following contents after step 010:

[0101] Step 011: If the target user's biological collection data is image data, determine whether the image data meets the preset biometric image quality requirements. If so, pre-process the target user's biological collection data to obtain the target user's biological image data.

[0102] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can effectively improve the application breadth of data collection by providing a method for processing biological image data, and can perform reliability and accuracy of biometric recognition according to the target biometric recognition algorithm.

[0103] In order to effectively improve the application of data collection, in one embodiment of the biometric recognition method provided in this application, see Figure 7 The biometric feature recognition method further includes the following contents after step 010:

[0104] Step 012: If the biological data collected by the target user is video data, frame extraction is performed on the video data based on a preset frame extraction rule to obtain a plurality of image data corresponding to the target user.

[0105] Step 013: Determine whether there is image data that meets the preset biometric image quality requirements among the image data. If so, pre-process the image data that meets the biometric image quality requirements to obtain the biometric image data of the target user.

[0106] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can effectively improve the application breadth of data collection by providing a processing method for biological video data, and can perform reliability and accuracy of biometric recognition according to the target biometric recognition algorithm.

[0107] In order to effectively improve the application of data collection, in one embodiment of the biometric recognition method provided in this application, see Figure 8 The biometric feature recognition method further includes the following contents after step 200:

[0108] Step 300: Obtain the biometric processing type corresponding to the biometric identification request.

[0109] Step 400: Process the target biometric feature recognition result corresponding to the biometric image data of the target user based on the biometric feature processing type, and output the corresponding processing result.

[0110] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can improve the applicability of the biometric recognition method by processing the target biometric recognition result corresponding to the biometric image data of the target user based on the biometric processing type. It is particularly suitable for scenarios where financial institutions perform subsequent transaction processing after performing biometric recognition on users during user transactions, and can effectively improve the security and reliability of processes such as subsequent transaction processing.

[0111] In order to further improve the applicability and comprehensiveness of the biometric identification method, in one embodiment of the biometric identification method provided in the present application, the biometric processing type includes: biometric registration; see Fig. 9 , step 400 in the biometric identification method specifically includes the following contents:

[0112] Step 411: storing the target biometric feature recognition result corresponding to the biometric image data of the target user into at least one biometric feature recognition result corresponding to the identifier of the target user to complete the biometric feature registration for the target biometric feature recognition result.

[0113] Step 412: Output the processing result of biometric registration of the target biometric recognition result.

[0114] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can further improve the applicability and comprehensiveness of the biometric recognition method by performing biometric registration on the corresponding target biometric recognition results, and is particularly suitable for biometric registration scenarios.

[0115] In order to further improve the applicability and comprehensiveness of the biometric identification method, in one embodiment of the biometric identification method provided in the present application, the biometric processing type includes: one-to-one identification (also referred to as 1:1 identification); Fig.10 , step 400 in the biometric identification method specifically includes the following contents:

[0116] Step 421: Perform one-to-one identification on a target biometric feature recognition result corresponding to the biometric image data of the target user and a pre-stored biometric feature recognition result corresponding to the identifier of the target user to obtain a corresponding one-to-one recognition result.

[0117] Step 422: Output the one-to-one recognition result.

[0118] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can further improve the applicability and comprehensiveness of the biometric recognition method by performing 1:1 recognition on the corresponding target biometric recognition results, and is particularly suitable for 1:1 recognition scenarios.

[0119] In order to further improve the applicability and comprehensiveness of the biometric identification method, in one embodiment of the biometric identification method provided in the present application, the biometric processing type includes: one-to-many identification (also referred to as 1:n identification); Fig.11 , step 400 in the biometric identification method specifically includes the following contents:

[0120] Step 431: Perform one-to-many recognition on the target biometric feature recognition result corresponding to the biometric image data of the target user and the pre-stored multiple biometric feature recognition results corresponding to the identifier of the target user to obtain a corresponding one-to-many recognition result.

[0121] Step 432: Output the one-to-many recognition result.

[0122] From the above description, it can be seen that the biometric recognition method provided in the embodiment of the present application can further improve the applicability and comprehensiveness of the biometric recognition method by performing 1:n recognition on the corresponding target biometric recognition results, and is particularly suitable for 1:n recognition scenarios.

[0123] From the software level, in order to solve the problems of existing multi-biometric recognition algorithm selection relying on manual work, poor reliability and low efficiency, the present application provides an embodiment of a biometric recognition device for executing all or part of the content of the biometric recognition method, see Fig.12 The biometric feature recognition device specifically includes the following contents:

[0124] The decision module 10 is used to input the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and select one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model.

[0125] In the decision module 10, the biometric identification request can be sent by the technical staff of the financial institution or the financial user through the client device held by the financial user to the biometric identification device. It can be understood that the biometrics mentioned in one or more embodiments of the present application refer to the physiological characteristics (fingerprints, irises, facial features, etc.) or behavioral characteristics (gait, etc.) inherent in the human body.

[0126] The recognition module 20 is used to obtain a target biometric feature recognition result corresponding to the biometric image data of the target user based on the target biometric feature recognition algorithm.

[0127] In the identification module 20, biometric features refer in particular to feature data that can be collected through images or videos. For example, the biometric image data can be facial image data, iris image data, retinal image data, facial image data, vein distribution map, three-dimensional palm image data, and signature image data of the target user; if the collected data is video data, the target user corresponds to multiple biometric image data obtained by extracting key frames from the video data.

[0128] The embodiments of the biometric identification device provided in the present application can be specifically used to execute the processing flow of the embodiments of the biometric identification method in the above embodiments. Its functions will not be described in detail here, and reference can be made to the detailed description of the above method embodiments.

[0129] From the above description, it can be seen that the biometric recognition device provided in the embodiment of the present application, by inputting the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and selecting one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model, can effectively realize the dynamic decision of biometric multi-algorithms integrating scene characteristics without human participation, make up for the defect that traditional biometric systems can only use a single algorithm for recognition, and provide an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the system generalization ability; it can effectively reduce the cost of manual adjustment when applying multiple algorithms in emerging biometric systems, greatly reduce the risks of easy errors and impact on recognition pass rate caused by manual adjustments, and improve the intelligence level of biometric systems.

[0130] In order to further illustrate the present scheme, the present application provides a biometric identification method with multi-algorithm intelligent decision-making implemented by a biometric identification system using multi-algorithm intelligent decision-making (i.e., a specific example of the aforementioned biometric feature identification method). A biometric multi-algorithm intelligent decision-making model is constructed based on the artificial intelligence technology deep neural network DNN to realize dynamic switching of the optimal algorithm in the transaction dimension. Model features are constructed based on the three dimensions of customer portrait, business characteristics, and algorithm capabilities; sample data comes from business data of various biometric scenarios, and the annotation method uses multiple algorithms for voting through background batch jobs, and combines customer portraits and business requirements to evaluate the optimal algorithm for the transaction; verification samples also come from biometric business data. This application example can well solve the dynamic decision-making of biometric multi-algorithms that integrate scene characteristics, and can automatically select the optimal algorithm for each biometric transaction without human participation, supporting the needs of multi-algorithm intelligent applications in various scenarios in the financial field.

[0131] In view of the defects and deficiencies in the existing technology, this application example provides a biometric system and method with multi-algorithm intelligent decision-making. Based on artificial intelligence technology, a biometric multi-algorithm intelligent decision-making model is constructed to automatically select the optimal algorithm for each biometric transaction for identification, solving the current problem of manual adjustment of multiple algorithms based on expert rules. On the one hand, this application example makes up for the defect that traditional biometric systems can only use a single algorithm for identification, and provides an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the generalization ability of the system; on the other hand, it reduces the cost of manual adjustment when applying multiple algorithms in emerging biometric systems, greatly reduces the risk of errors caused by manual adjustments and affecting the recognition pass rate, and improves the intelligence level of biometric systems.

[0132] This application example provides a biometric identification system and method for multi-algorithm intelligent decision-making, wherein the biometric identification system for multi-algorithm intelligent decision-making includes a biometric identification main control module, a biometric feature acquisition module, a data transmission module, and a biometric identification multi-algorithm intelligent decision-making module. When performing a multi-algorithm biometric identification transaction, the biometric feature acquisition module drives the camera to collect the user's biometric image or video unstructured data, and then uploads the biometric image or video unstructured data to the biometric identification multi-algorithm intelligent decision-making module through the data transmission module for algorithm scheduling and decision-making. After completing the algorithm selection, the best algorithm is used to perform at least one of the related biometric feature registration, 1:1 recognition and 1:n recognition algorithms. After the algorithm processing is completed, the processing result is sent to the intelligent interactive device end.

[0133] The biometric acquisition module mainly completes the acquisition of non-structured data such as biological images or videos and data quality detection and control functions. The acquisition control unit drives the camera device deployed on the intelligent interactive device to collect the user's biological image or video unstructured data; the biometric acquisition main control unit sends the collected biological image or video unstructured data to the biological quality control unit. When the collected data is an image, it is directly judged whether the organism in the picture meets the use standard: such as the interpupillary distance of the human eye>=60 pixels, whether there is motion blur, the image is too bright, the image is too dark, there is no organism, etc.; when the collected data is a video, the video data is first extracted for key frames. The video requires a duration of less than 5 seconds. In order to improve efficiency, 2 frames or s are extracted as key frame detection to determine whether the biological features in the image frame meet the use standard: such as the interpupillary distance>=60 pixels, whether the image is motion blur, the image is too bright, the image is too dark, there is no biological feature, etc. If it meets the requirements, a biometric data transmission processing request is initiated to complete the biometric data acquisition process.

[0134] The data transmission module mainly completes the upload of biometric image or video unstructured data and the distribution of biometric pictures or features. The biometric data transmission main control unit first performs data security check control to determine whether the uploaded data is biometric image or video unstructured data. If so, the biometric data is uploaded to the cloud database via http or socket; when the data is downloaded, the download unit directly initiates an image or feature query application to the cloud database, and after the corresponding results are queried, the data download unit sends them to the intelligent interactive device or application server, thereby completing the biometric data transmission processing flow.

[0135] The biometric multi-algorithm intelligent decision module mainly completes the scheduling of biometric multi-algorithms, multi-algorithm intelligent selection, biometric registration and biometric identification processing, and finally returns the relevant registration or identification results to the intelligent device processing front end. First, the biometric multi-algorithm scheduling is performed, the type and number of server-side algorithm models are retrieved, and the scheduling results are sent to the multi-algorithm decision model for the best algorithm selection. Different versions of biometric algorithms from different manufacturers have their own advantages and disadvantages in specific scenarios. At present, the selection of biometric multi-algorithms is manually adjusted based on expert rules, and it is impossible to integrate the scene characteristics for optimal application. The main function of the biometric multi-algorithm decision model is to integrate the scene and application characteristics to achieve the optimal algorithm selection based on channels and transactions. The multi-algorithm decision model is trained with a deep learning network DNN, and the model features are mainly constructed based on three feature dimensions: customer portrait, business requirements, and algorithm capabilities. The sample data is comprehensively derived from the business data of various biological scenarios. Data annotation is achieved through the background batch job using multiple algorithms for automatic voting and evaluation, and the verification sample comes from the biometric business scenario data. Before algorithm processing, first determine whether the processed data is an image. If it is an image, it is directly sent to the corresponding algorithm processing unit. If it is a video, first extract the best biometric data frame and then send it to the corresponding algorithm processing unit. Send the image data to the corresponding algorithm service processing unit for processing: For data that needs to be processed by the biometric registration algorithm, perform data preprocessing, complete biometric extraction and modeling, and then register the user information and biometrics to the biometric system database. For requests that require 1:1 biometric algorithm processing, perform image preprocessing, complete biometric extraction and modeling, and then perform a 1:1 comparison of the feature data with the biometrics queried in the database, and return the comparison results to the requesting client. For those that need to be processed by a 1:n biometric algorithm, complete image preprocessing, complete biometric extraction and modeling, and perform 1:n recognition with N biometrics in the database, and return the recognition results to the requesting client. Among them, N and n mentioned in one or more embodiments of the present application are positive integers.

[0136] Based on the above content, the multi-algorithm intelligent decision-making biometric recognition system is as follows Fig.13 As shown, it includes: a biometric identification main control module 1, a biometric feature collection module 2, a data transmission module 3 and a biometric identification multi-algorithm intelligent decision module 4. The biometric identification main control module 1, the biometric feature collection module 2, the data transmission module 3 and the biometric identification multi-algorithm intelligent decision module 4 communicate with each other through HTTPS.

[0137] Among them, see Fig.14The biometric collection module 2 is used for the overall control of the front end of biometric data collection, including a biometric collection main control MCU unit 21, a biometric non-structured data collection control unit 22 and a biometric non-structured data quality control unit 23. The communication between the biometric collection main control MCU unit 21, the biometric non-structured data collection control unit 22 and the biometric non-structured data quality control unit 2 is completed through JavaScript instant messaging or Java internal communication. The front-end biometric collection main control module is mainly deployed in the intelligent interactive device terminal.

[0138] See also Fig.15 The data transmission module 3 is used for uploading and downloading biometric data. It includes a biometric unstructured data transmission main control unit 31, a data uploading unit 32 and a data sending unit 33. The communication between the units is completed through JavaScript instant messaging or Java internal communication. It is mainly deployed in the intelligent interactive device terminal.

[0139] See also Fig.16 , the biometric multi-algorithm intelligent decision module 4 is used to complete the biometric multi-algorithm intelligent decision and algorithm processing functions. It includes a biometric algorithm main control unit 41 and a biometric algorithm intelligent decision unit 42; the biometric algorithm intelligent decision unit 42 includes: a biometric multi-algorithm scheduling unit 421, a biometric multi-algorithm model decision unit 422, an algorithm A (biological A algorithm subunit) 423, an algorithm B (biological B algorithm subunit) 424 and an algorithm C (biological C algorithm subunit) 425, etc., wherein algorithms A, B and C refer to different versions of biometric algorithm models from different manufacturers, respectively, and can be expanded horizontally and vertically, not limited to Fig.16 The multi-algorithm intelligent decision module 4 also includes a biometric registration algorithm processing unit 43, a feature 1:1 recognition algorithm processing unit 44 and a feature 1:n recognition algorithm processing unit 45. The units communicate with each other through Java internal communication, and the modules are mainly deployed on the background server.

[0140] See also Fig.17 , the execution process of the biometric main control module in this application example is as follows:

[0141] Step S101: Initiate a biometric data collection request;

[0142] Step S102: According to the data collection instruction, the front-end camera is driven to collect biometric image or video data;

[0143] Step S103: sending the biometric data to the quality judgement device;

[0144] Step S104: Determine whether the captured image is a biometric image. If the captured image is a biometric image, proceed to step S107. If the captured image is not a biometric image, proceed to step S105.

[0145] Step S105: when it is determined in step S104 that the collected data is not a biometric image, determining whether the collected data is biometric video data;

[0146] Step S106: when it is determined by step S105 that the collected data is not biometric video data, return to step S102 for re-collection; when it is determined by step S105 that the collected data is biometric video data, extract the video key frames, enter step S107, and detect frame by frame;

[0147] Step S107: Perform biometric image quality determination, such as whether the interpupillary distance of human eyes is greater than or equal to 60 pixels, whether there is motion blur, the image is too bright, the image is too dark, there is no living creature, etc. If so, return to step S102 for re-collection;

[0148] Step S108: Initiate a request to upload biometric data;

[0149] Step S109: biometric data collection is completed;

[0150] Step S110: The biometric data collection process ends.

[0151] See also Fig.18 ,The upload execution process of the data transmission module in this application example is as follows:

[0152] Step S201: Initiate a biometric data upload request;

[0153] Step S202: Performing biometric data security access control;

[0154] Step S203: Determine whether the uploaded data is biometric image data or video data, if so proceed to step S206, if not proceed to step S204;

[0155] Step S204: biometric data upload request rejected;

[0156] Step S205: the biometric data upload process ends;

[0157] Step S206: when it is determined in step S204 that the uploaded data is a biometric image or video data, it is determined whether to upload a biometric registration image or video data;

[0158] Step S207: when it is determined in step S206 that the image data or video data to be uploaded is registered as a biometric feature, the data is uploaded to the cloud database via https or socket;

[0159] Step S208: Initiate a biometric algorithm intelligent decision request;

[0160] Step S209: biometric registration data upload is completed;

[0161] Step S210: the biometric data upload process ends;

[0162] Step S211: when it is determined in step S206 that the biometric registration image or video data is not uploaded, determine whether to upload the biometric 1:1 recognition image data or video data;

[0163] Step S212: when it is determined in step S211 that the biological 1:1 recognition image data or video data is to be uploaded, the data is uploaded to the cloud database via https or socket;

[0164] Step S213: Initiate a biometric algorithm intelligent decision and 1:1 recognition algorithm processing request;

[0165] Step S214: biological 1:1 data upload completed;

[0166] Step S215: the biometric data upload process ends;

[0167] Step S216: when it is determined in step S211 that the biological 1:1 recognition image data or video data is not uploaded, it is determined whether the biological 1:n recognition image data or video data is uploaded;

[0168] Step S217: when it is determined in step S216 that the biological 1:n recognition image data or video data is to be uploaded, the data is uploaded to the cloud database via https or socket;

[0169] Step S218: Initiate a biometric algorithm intelligent decision and 1:n recognition algorithm processing request;

[0170] Step S219: biological 1:n data upload completed;

[0171] Step S220: the biometric data upload process ends;

[0172] Step S221: when it is determined in step S216 that the biometric data upload request is not for uploading biological 1:n recognition image data or video data, the biometric data upload request is rejected;

[0173] Step S222: The biometric data upload process ends.

[0174] See also Fig.19 , the execution process of the data transmission module in this application example is as follows:

[0175] Step S301: Initiate a request to send biometric data;

[0176] Step S302: determining whether the sent data is biometric data or image data;

[0177] Step S303: when it is determined in step 301 that the biometric data is not to be downloaded or the image data is not to be downloaded, the request for downloading the biometric data is rejected;

[0178] Step S304: the biometric data delivery process ends;

[0179] Step S305: when it is determined in step S302 that the data to be sent is biometric feature data or image data, it is determined whether the data to be sent is 1:1 biological recognition feature data or image data;

[0180] Step S306: when it is determined in step S305 that the downloaded biological 1:1 feature data or image data is necessary, query one biological feature data or image data from the cloud database according to the customer information;

[0181] Step S307: Send one biometric feature or image to a designated interactive device or application server;

[0182] Step S308: The request for sending biometric data is completed;

[0183] Step S309: the biometric data delivery process ends;

[0184] Step S310: when it is determined in step S305 that the 1:1 identification feature data or image data of the organism is not generated, it is determined whether the 1:n identification feature data or image data of the organism is generated;

[0185] Step S311: when it is determined in step S310 that the following biological 1:n identification feature data or image data is generated, query and obtain N biological feature data or image data from the cloud database according to the customer grouping information;

[0186] Step S312: Sending N biometric features or images to a designated interactive device or application server;

[0187] Step S313: The request for sending biometric data is completed;

[0188] Step S314: the biometric data delivery process ends;

[0189] Step S315: when it is determined in step S310 that the biometric data is not a 1:n identification feature or image data, the request for sending the biometric data is rejected;

[0190] Step S316: The biometric data sending process ends.

[0191] See also Fig. 20 ,The execution process of the biometric multi-algorithm intelligent decision-making module in this application example is as follows:

[0192] Step S401: Initiate a biometric algorithm processing request;

[0193] Step S402: selecting weights according to a customer portrait calculation algorithm;

[0194] Step S403: Calculate the algorithm selection weight according to the scenario business requirements;

[0195] Step S404: Calculate the algorithm selection weight according to the algorithm capability;

[0196] Step S405: performing biometric multi-algorithm scheduling;

[0197] Step S406: calling the biometric multi-algorithm decision model;

[0198] Step S407: Based on the above weights, the multi-algorithm decision model selects the optimal processing algorithm for the transaction;

[0199] Step S408: determining whether the processed data is an image. If it is image data, proceeding to step S413 for processing;

[0200] Step S409: when it is determined in step S408 that the processed data is not an image, determining whether the processed data is a video;

[0201] Step S410: when it is determined in step S410 that the processed data is video data, the best video frame is extracted from the video data, and the process proceeds to step S413 for processing;

[0202] Step S411: when it is determined in step S409 that the processed data is not video data, the biometric algorithm processing request is rejected;

[0203] Step S412: the biometric algorithm processing flow ends;

[0204] Step S413: Determine whether to perform a biometric registration algorithm processing request;

[0205] Step S414: when it is determined in step S413 that a biometric registration algorithm processing request is to be made, the request is sent to the biometric registration algorithm processing unit;

[0206] Step S415: preprocessing biological image or video unstructured data using the optimal algorithm selected by the decision model;

[0207] Step S416: extracting and modeling biometric features using the optimal algorithm selected by the decision model;

[0208] Step S417: Save the biometric features to a cloud database;

[0209] Step S418: Return the registration result to the requesting client;

[0210] Step S419: the biometric registration algorithm processing request is completed;

[0211] Step S420: the biometric algorithm processing flow ends;

[0212] Step S421: when it is determined in step S413 that the biometric registration algorithm processing is not being performed, determining whether a biometric 1:1 recognition algorithm processing request is being performed;

[0213] Step S422: When it is determined in step S421 that a biological 1:1 recognition algorithm processing request is to be performed, the request is sent to the biological 1:1 recognition algorithm processing unit;

[0214] Step S423: preprocessing biological image or video unstructured data using the optimal algorithm selected by the decision model;

[0215] Step S424: extracting and modeling biometric features using the optimal algorithm selected by the decision model;

[0216] Step S425: Perform a 1:1 comparison with a biometric feature acquired from the cloud;

[0217] Step S426: Return the comparison result to the requesting client

[0218] Step S427: Biometric 1:1 recognition algorithm processing request completed

[0219] Step S428: the biometric algorithm processing flow ends;

[0220] Step S429: when it is determined in step S421 that the request for processing the biological 1:1 recognition algorithm is not being made, determining whether a request for processing the biological 1:n recognition algorithm is being made;

[0221] Step S430: When it is determined in step S429 that a biological 1:n recognition algorithm processing request is to be performed, the request is sent to the biological 1:n recognition algorithm processing unit;

[0222] Step S431: preprocessing biological image or video unstructured data using the optimal algorithm selected by the decision model;

[0223] Step S432: extracting and modeling biometric features using the optimal algorithm selected by the decision model;

[0224] Step S433: Perform 1:n recognition with the N biometric features obtained from the cloud, and return the top 1 customer information as the recognition result;

[0225] Step S434: Return the recognition result to the requesting client;

[0226] Step S435: biological 1:n recognition algorithm processing request is completed;

[0227] Step S436: the biometric algorithm processing flow ends;

[0228] Step S437: when it is determined in step S429 that the request is not for biometric 1:n recognition algorithm processing, the biometric recognition algorithm processing request is rejected;

[0229] Step S438: The biometric algorithm processing flow ends.

[0230] From the above description, it can be seen that the biometric system with multi-algorithm intelligent decision-making and the biometric method with multi-algorithm intelligent decision-making provided by the application example of this application use artificial intelligence technology and deep neural network DNN training to build a biometric multi-algorithm intelligent decision-making model, which can automatically select the optimal algorithm for each biometric transaction data for identification. On the one hand, this application example makes up for the defect that the traditional biometric system can only use a single algorithm for identification, and provides an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the generalization ability of the system; on the other hand, it reduces the cost of manual adjustment when applying multiple algorithms in emerging biometric systems, greatly reduces the risk of errors caused by adjustments, affects the recognition pass rate, and improves the intelligence level of the biometric system.

[0231] From the hardware level, in order to solve the problems that the existing multi-biometric recognition algorithm selection relies on manual work, has poor reliability and low efficiency, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the biometric feature recognition method, and the electronic device specifically includes the following contents:

[0232] Fig.21 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig.21 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig.21 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0233] In one embodiment, the biometric feature recognition function may be integrated into a central processing unit, wherein the central processing unit may be configured to perform the following control:

[0234] Step 100: Input the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and select one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user based on the output of the multi-algorithm decision model.

[0235] Step 200: Obtain a target biometric feature recognition result corresponding to the biometric image data of the target user based on the target biometric feature recognition algorithm.

[0236] From the above description, it can be seen that the electronic device provided in the embodiment of the present application, by inputting the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and selecting one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model, can effectively realize the dynamic decision of biometric multi-algorithms integrating scene characteristics without human participation, make up for the defect that the traditional biometric recognition system can only use a single algorithm for recognition, and provide an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the system generalization ability; it can effectively reduce the cost of manual adjustment when applying multiple algorithms in emerging biometric recognition systems, greatly reduce the risks of easy errors and impact on recognition pass rate caused by manual adjustments, and improve the intelligence level of the biometric recognition system.

[0237] In another embodiment, the biometric identification device can be configured separately from the central processing unit 9100. For example, the biometric identification device can be configured as a chip connected to the central processing unit 9100, and the biometric identification function is realized through the control of the central processing unit.

[0238] like Fig.21 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig.21 In addition, the electronic device 9600 may also include Fig.21 For components not shown, reference may be made to the prior art.

[0239] like Fig.21 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0240] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0241] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0242] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0243] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0244] The communication module 9110 is a transmitter / receiver 9110 that sends and receives signals via an antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0245] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module, etc. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0246] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the biometric feature recognition method in the above embodiments. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, all the steps of the biometric feature recognition method in the above embodiments are implemented by the execution subject being a server or a client. For example, when the processor executes the computer program, the following steps are implemented:

[0247] Step 100: Input the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and select one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user based on the output of the multi-algorithm decision model.

[0248] Step 200: Obtain a target biometric feature recognition result corresponding to the biometric image data of the target user based on the target biometric feature recognition algorithm.

[0249] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application, by inputting the environmental feature data corresponding to the target user who initiates the biometric recognition request into a preset multi-algorithm decision model, and selecting one of multiple preset biometric recognition algorithms as the target biometric recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model, can effectively realize the dynamic decision-making of biometric multi-algorithms integrating scene characteristics without human participation, make up for the defect that traditional biometric systems can only use a single algorithm for recognition, and provide an intelligent decision-making solution for multiple algorithms to make decisions at the same time, greatly improving the generalization ability of the system; it can effectively reduce the cost of manual adjustment when applying multiple algorithms in emerging biometric systems, greatly reduce the risks of easy errors and impact on recognition pass rate caused by manual adjustments, and improve the intelligence level of biometric systems.

[0250] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (devices), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the functions in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0251] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0252] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A biometric identification method, characterized in that: include: Inputting the environmental feature data corresponding to the target user who initiates the biometric feature recognition request into a preset multi-algorithm decision model, and selecting one of the multiple preset biometric feature recognition algorithms as the target biometric feature recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model; Acquire a target biometric feature recognition result corresponding to the biometric image data of the target user based on the target biometric feature recognition algorithm; The environmental feature data includes: customer profile data, business requirement data and algorithm capability data; Correspondingly, before inputting the environmental feature data corresponding to the target user initiating the biometric feature recognition request into the preset multi-algorithm decision model, it also includes: Receiving a biometric identification request and corresponding biometric data of a target user, wherein the biometric identification request includes a unique identifier of the target user; Based on the unique identifier of the target user, the customer portrait data corresponding to the target user is obtained, and the pre-stored business requirement data and algorithm capability data are retrieved.

2. The biometric feature recognition method according to claim 1, characterized in that: The step of inputting the environmental feature data corresponding to the target user initiating the biometric feature recognition request into a preset multi-algorithm decision model, and selecting one of a plurality of preset biometric feature recognition algorithms as the target biometric feature recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model, comprises: Acquire the target algorithm identifier output by the multi-algorithm decision model according to the customer profile data, business requirement data and algorithm capability data corresponding to the target user; The biometric recognition algorithm corresponding to the target algorithm identifier is determined from the one-to-one correspondence between each pre-stored algorithm identifier and each biometric recognition algorithm, and the biometric recognition algorithm is determined as the target biometric recognition algorithm corresponding to the target user.

3. The biometric feature recognition method according to claim 2, characterized in that: The environmental feature data also includes: weights corresponding to customer profile data, business requirement data, and algorithm capability data; Correspondingly, the step of obtaining the target algorithm identifier output by the multi-algorithm decision model according to the customer profile data, business requirement data, and algorithm capability data corresponding to the target user includes: The customer portrait data, business requirement data and algorithm capability data corresponding to the target user, as well as the weights corresponding to the customer portrait data, business requirement data and algorithm capability data corresponding to the target user are input into the multi-algorithm decision model so that the multi-algorithm decision model outputs a target algorithm identifier.

4. The biometric feature recognition method according to claim 1, characterized in that: Also includes: Obtaining a training data set, wherein the training data set includes historical environmental feature data of multiple users and labels corresponding to the historical environmental feature data of each user, wherein the label is one of algorithm identifiers corresponding to each preset biometric feature recognition algorithm; The training data set is trained with a deep learning network algorithm to obtain a multi-algorithm decision model for selecting among the various biometric feature recognition algorithms and outputting corresponding algorithm identifiers.

5. The biometric feature recognition method according to claim 4, characterized in that: The historical environment characteristic data includes: customer portrait data, business requirement data and algorithm capability data, and the historical environment characteristic data also includes: the weights corresponding to the customer portrait data, business requirement data and algorithm capability data.

6. The biometric feature recognition method according to claim 1, characterized in that: After receiving the biometric feature recognition request and the corresponding biometric collection data of the target user, the method further includes: If the target user's biological data is image data, it is determined whether the image data meets the preset biometric image quality requirements. If so, the target user's biological data is preprocessed to obtain the target user's biological image data.

7. The biometric feature recognition method according to claim 1, characterized in that: After receiving the biometric feature recognition request and the corresponding biometric collection data of the target user, the method further includes: If the biological data collected by the target user is video data, performing frame extraction processing on the video data based on a preset frame extraction rule to obtain a plurality of image data corresponding to the target user; It is determined whether there is image data that meets the preset biometric image quality requirements among the image data. If so, the image data that meets the biometric image quality requirements are preprocessed to obtain the biometric image data of the target user.

8. The biometric feature recognition method according to any one of claims 1 to 7, characterized in that: Also includes: Obtaining a biometric processing type corresponding to the biometric identification request; The target biometric feature recognition result corresponding to the biometric image data of the target user is processed based on the biometric feature processing type, and the corresponding processing result is output.

9. The biometric feature recognition method according to claim 8, characterized in that: The biometric processing types include: biometric registration; Correspondingly, the target biometric feature recognition result corresponding to the biometric image data of the target user is processed based on the biometric feature processing type, including: The target biometric feature recognition result corresponding to the biometric image data of the target user is stored in at least one biometric feature recognition result corresponding to the identifier of the target user to complete the biometric feature registration for the target biometric feature recognition result.

10. The biometric feature recognition method according to claim 8, characterized in that: The biometric processing types include: one-to-one identification; A target biometric feature recognition result corresponding to the biometric image data of the target user is identified one-to-one with a pre-stored biometric feature recognition result corresponding to the identifier of the target user to obtain a corresponding one-to-one recognition result.

11. The biometric feature recognition method according to claim 8, characterized in that: The biometric processing types include: one-to-many identification; A target biometric feature recognition result corresponding to the biometric image data of the target user is subjected to one-to-many recognition with a plurality of pre-stored biometric feature recognition results corresponding to the identifier of the target user to obtain a corresponding one-to-many recognition result.

12. A biometric identification device, characterized in that: include: A decision module, used to input the environmental feature data corresponding to the target user who initiates the biometric feature recognition request into a preset multi-algorithm decision model, and select one of the multiple preset biometric feature recognition algorithms as the target biometric feature recognition algorithm corresponding to the target user according to the output of the multi-algorithm decision model; An identification module, used for obtaining a target biometric identification result corresponding to the biometric image data of the target user based on the target biometric identification algorithm; The environmental feature data includes: customer profile data, business requirement data and algorithm capability data; Correspondingly, before inputting the environmental feature data corresponding to the target user initiating the biometric feature recognition request into the preset multi-algorithm decision model, it also includes: Receiving a biometric identification request and corresponding biometric data of a target user, wherein the biometric identification request includes a unique identifier of the target user; Based on the unique identifier of the target user, the customer portrait data corresponding to the target user is obtained, and the pre-stored business requirement data and algorithm capability data are retrieved.

13. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the biometric feature recognition method according to any one of claims 1 to 11 is implemented.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the biometric feature recognition method according to any one of claims 1 to 11 is implemented.

Images