Face recognition method, device and system

Abstract

The embodiments of the present application provide a face recognition method, device and system, which can also be used in the financial field. The method includes: capturing a face image and extracting a first face feature according to a first image recognition algorithm, and storing the face feature in a local feature queue; performing feature matching on the face features in the local feature queue and a local target feature library, and sending the feature matching result, the face image and the first face feature to a face recognition platform, so that the face recognition platform extracts a second face feature corresponding to the face image according to a second image recognition algorithm when the feature matching result is passed, and performs feature matching on the second face feature and the local target feature library to obtain a face recognition result; the present application can effectively save server resources and improve face recognition efficiency and accuracy.

Description

Technical Field

[0001] The present application relates to the field of biometrics and can also be used in the financial field, and specifically relates to a face recognition method, device and system. Background Art

[0002] At present, face recognition systems have been widely used in the industry. The common usage is to obtain face images through the camera at the front end, and then transmit the face images to the face recognition system on the backend server for face recognition. Generally speaking, face recognition systems mainly include two usage scenarios: 1:1 face authentication and 1:N face search. For 1:N face search, as N increases, the search speed will decrease, resulting in an increase in search time. Therefore, the common practice is to control N within a certain range, and support 1:N face search under massive data by horizontally expanding the capacity of server nodes and adopting parallel search.

[0003] The inventors have discovered that in the prior art, due to the limited resources of a single server, in order to support face searches under massive data, a large number of server nodes need to be deployed, which easily leads to a waste of resources. Summary of the invention

[0004] In response to the problems in the prior art, the present application provides a face recognition method, device and system, which can effectively save server resources and improve the efficiency and accuracy of face recognition.

[0005] In order to solve at least one of the above problems, the present application provides the following technical solutions:

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

[0007] Capturing a facial image and extracting a first facial feature according to a first image recognition algorithm, and storing the facial feature in a local feature queue;

[0008] Feature matching is performed on the facial features in the local feature queue and the local target feature library, and the feature matching result, the facial image and the first facial feature are sent to the facial recognition platform, wherein the local target feature library stores the target facial feature value sent by the facial recognition platform, so that the facial recognition platform extracts the second facial feature corresponding to the facial image according to the second image recognition algorithm when the feature matching result is passed, and feature matching is performed on the second facial feature and the local target feature library to obtain the facial recognition result, and the facial recognition platform sends the first facial feature to the local feature queue of the corresponding facial capture camera when the feature matching result is failed.

[0009] Furthermore, after storing the facial features in the local feature queue, the method further includes:

[0010] Determine whether the local feature queue exceeds a feature quantity threshold or whether the current CPU usage exceeds an operation threshold;

[0011] If so, the feature matching operation is terminated and the facial image and the first facial feature are sent to the facial recognition platform.

[0012] In a second aspect, the present application provides a face recognition method, comprising:

[0013] Receive a face image, a first face feature, and a feature matching result sent by at least one face capture camera, wherein the first face feature is extracted according to a first image recognition algorithm after the face capture camera captures the face image, and the feature matching result is obtained by the face capture camera storing the face feature in a local feature queue and then performing feature matching on the face features in the local feature queue with a local target feature library, and the local target feature library of the face capture camera stores target face feature values ​​sent by a face recognition platform;

[0014] Determine whether the feature matching result passes. If so, extract the second facial feature corresponding to the facial image according to the second image recognition algorithm, and perform feature matching on the second facial feature with the local target feature library to obtain the facial recognition result. Otherwise, send the first facial feature to the local feature queue of the corresponding facial capture camera.

[0015] Further, extracting a second facial feature corresponding to the facial image according to a second image recognition algorithm, and performing feature matching on the second facial feature with a local target feature library to obtain a facial recognition result includes:

[0016] Extracting features from the face image according to the target person information to be confirmed in the matching result, the face image corresponding to the target person information to be confirmed, and a second image recognition algorithm to obtain a second face feature;

[0017] It is determined whether the second facial feature matches the local target feature library. If so, a corresponding face recognition result is obtained. Otherwise, the first facial feature is sent to the local feature queue of the corresponding face capture camera.

[0018] In a third aspect, the present application provides a face recognition device, comprising:

[0019] A face capture camera image analysis module, used to capture a face image and extract a first face feature according to a first image recognition algorithm, and store the face feature in a local feature queue;

[0020] A face capture camera information uploading module is used to perform feature matching on the face features in the local feature queue and the local target feature library, and send the feature matching result, the face image and the first face feature to the face recognition platform, wherein the local target feature library stores the target face feature value sent by the face recognition platform, so that the face recognition platform extracts the second face feature corresponding to the face image according to the second image recognition algorithm when the feature matching result is passed, and performs feature matching on the second face feature and the local target feature library to obtain the face recognition result, and the face recognition platform sends the first face feature to the local feature queue of the corresponding face capture camera when the feature matching result is failed.

[0021] Furthermore, it also includes:

[0022] A face capture camera performance monitoring unit, used to determine whether the local feature queue exceeds a feature quantity threshold or whether the current CPU usage exceeds an operating threshold;

[0023] A performance abnormality sending unit is used to terminate the feature matching operation and send the face image and the first face feature to the face recognition platform if the local feature queue exceeds the feature quantity threshold or the current CPU usage exceeds the operation threshold.

[0024] In a fourth aspect, the present application provides a face recognition device, comprising:

[0025] A face recognition platform image analysis module, used for receiving a face image, a first face feature and a feature matching result sent by at least one face capture camera, wherein the first face feature is extracted according to a first image recognition algorithm after the face capture camera captures the face image, and the feature matching result is obtained by the face capture camera storing the face feature in a local feature queue and then performing feature matching on the face features in the local feature queue with a local target feature library, and the local target feature library of the face capture camera stores target face feature values ​​sent by the face recognition platform;

[0026] The face recognition platform information processing module is used to determine whether the feature matching result is passed. If so, the second face feature corresponding to the face image is extracted according to the second image recognition algorithm, and the second face feature is feature matched with the local target feature library to obtain the face recognition result. Otherwise, the first face feature is sent to the local feature queue of the corresponding face capture camera.

[0027] In a fifth aspect, the present application provides a face recognition system, comprising: a face capture camera and a face recognition platform communicatively connected to the face capture camera;

[0028] The face capture camera comprises:

[0029] A face capture camera image analysis module, used to capture a face image and extract a first face feature according to a first image recognition algorithm, and store the face feature in a local feature queue;

[0030] A face capture camera information uploading module, used for matching the face features in the local feature queue with the local target feature library, and sending the feature matching result, the face image and the first face feature to the face recognition platform, wherein the local target feature library stores the target face feature value sent by the face recognition platform;

[0031] The face recognition platform includes:

[0032] A face recognition platform image analysis module, used for receiving a face image, a first face feature and a feature matching result sent by at least one face capture camera;

[0033] The face recognition platform information processing module is used to determine whether the feature matching result is passed. If so, the second face feature corresponding to the face image is extracted according to the second image recognition algorithm, and the second face feature is feature matched with the local target feature library to obtain the face recognition result. Otherwise, the first face feature is sent to the local feature queue of the corresponding face capture camera.

[0034] In a sixth 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 steps of the face recognition method when executing the program.

[0035] In a seventh aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the face recognition method when executed by a processor.

[0036] It can be seen from the above technical scheme that the present application provides a face recognition method, device and system, which, through the face capture camera's own function of supporting face search, forms a distributed network established by multiple face capture cameras, and uses the numerous face capture cameras on the distributed network to realize their respective face search functions, especially in the case of massive data, which can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms, can the target's identity information be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] 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.

[0038] Figure 1 This is one of the flowcharts of the face recognition method in the embodiment of the present application;

[0039] Figure 2 This is a second flow chart of the face recognition method in the embodiment of the present application;

[0040] Figure 3 This is the third flow chart of the face recognition method in the embodiment of the present application;

[0041] Figure 4 This is a fourth flow chart of the face recognition method in an embodiment of the present application;

[0042] Figure 5 This is one of the structural diagrams of the face recognition device in the embodiment of the present application;

[0043] Figure 6 This is the second structural diagram of the face recognition device in the embodiment of the present application;

[0044] Figure 7 This is the third structural diagram of the face recognition device in the embodiment of the present application;

[0045] Figure 8 A structural diagram of a face recognition system in an embodiment of the present application;

[0046] Fig. 9 This is a structural diagram of a face recognition system in a specific embodiment of the present application;

[0047] Fig.10 This is a flow chart of a face recognition system in a specific embodiment of the present application;

[0048] Fig.11 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0049] 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.

[0050] Taking into account the limited resources of a single server in the prior art, in order to support face search under massive data, the number of server nodes that need to be deployed will be very large, which will easily lead to the problem of resource waste. The present application provides a face recognition method, device and system. By utilizing the face search support function of the face capture camera itself, a distributed network established by multiple face capture cameras is formed, and the numerous face capture cameras on the distributed network are used to realize their respective face search functions. Especially in the case of massive data, server resources can be greatly saved. At the same time, different face recognition algorithms are applied through the combination of terminal face recognition and cloud face recognition to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined. This can effectively save server resources and improve the efficiency and accuracy of face recognition.

[0051] In order to effectively save server resources and improve the efficiency and accuracy of face recognition, the present application provides an embodiment of a face recognition method, the execution subject can be one or more face capture cameras, see Figure 1 , the face recognition method specifically includes the following contents:

[0052] Step S101: capturing a facial image and extracting a first facial feature according to a first image recognition algorithm, and storing the facial feature in a local feature queue.

[0053] Optionally, the executor of the method of the present application may be one or more face capture cameras, which can acquire images of objects captured by the cameras and detect images. Only when images with portraits are detected will subsequent processing procedures be triggered.

[0054] Specifically, facial features are extracted from the face image according to the corresponding first image recognition algorithm to obtain first facial features, and the first facial features are put into a local feature queue for subsequent face search.

[0055] Optionally, the local feature queue is used to store facial feature values, and the feature value data in the queue comes from two parts: 1. The first facial feature obtained by feature extraction; 2. The feature value sent from the face recognition platform.

[0056] Step S102: Feature matching is performed on the facial features in the local feature queue and the local target feature library, and the feature matching result, the facial image and the first facial feature are sent to the facial recognition platform, wherein the local target feature library stores the target facial feature value sent by the facial recognition platform, so that the facial recognition platform extracts the second facial feature corresponding to the facial image according to the second image recognition algorithm when the feature matching result is passed, and feature matches the second facial feature with the local target feature library to obtain the facial recognition result, and the facial recognition platform sends the first facial feature to the local feature queue of the corresponding facial capture camera when the feature matching result is failed.

[0057] Optionally, the local face search function of the applicant's face capture camera is used to read the records in the local feature queue, and implement the face search (i.e., feature matching) function locally on the face capture camera based on the terminal local target feature library, and send the feature matching result, the face image and the first face feature to the face recognition platform.

[0058] Optionally, the face recognition platform adopts a new second image recognition algorithm to achieve high-precision 1:N face search within a small data range, and returns the final face search result.

[0059] Specifically, the face recognition platform reads the matching results uploaded by the face capture camera from the message center. If the data is marked as "search failed", the process ends; if the data is marked as "search successful", a high-precision face recognition algorithm is used to perform a 1:N face search in a small range based on the target person information (several) to be confirmed in the result data and the initial target face image uploaded. Other face recognition algorithms different from those in the face capture camera are used to extract feature values ​​of the initial target face image. At the same time, based on the target person information to be confirmed, the face feature value data of the corresponding algorithm is obtained from the cloud face feature library, and the algorithm is used to perform a 1:N face search. The one with the highest similarity in the search result is the final target person information. Finally, the processing results are placed in the message center.

[0060] Optionally, the face recognition platform needs to complete face registration first, use an ordinary camera to obtain the user's own face image and upload it to the face recognition platform. The face recognition platform sends the face feature values ​​stored in the cloud to each face capture camera on the network according to predetermined rules. Each face capture camera only stores part of the face feature value data. When the subsequent user passes through the capture range of the face capture camera, the face capture camera captures the face image and extracts the face feature values, and performs a 1:N face search in the camera. If the current face capture camera can search for the target person who meets the threshold, the search result and the original target person image are sent to the face recognition platform; otherwise, the face feature values ​​to be searched are sent to the face recognition platform, and the face recognition platform sends the feature values ​​of the person to be searched to other face capture cameras on the network, and continues to perform a 1:N face search in other face capture cameras on the network.

[0061] From the above description, it can be seen that the face recognition method provided in the embodiment of the present application can form a distributed network established by multiple face capture cameras by utilizing the face search support function of the face capture camera itself, and utilize the numerous face capture cameras on the distributed network to realize their respective face search functions. Especially in the case of massive data, it can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0062] In order to ensure the reliability of face image recognition and avoid data loss or recognition delay caused by the performance problem of the face capture camera itself, in one embodiment of the face recognition method of the present application, see Figure 2 , after step S101, the following contents may also be specifically included:

[0063] Step S201: Determine whether the local feature queue exceeds a feature quantity threshold or whether the current CPU usage exceeds an operation threshold.

[0064] Step S202: If yes, terminate the feature matching operation and send the face image and the first face feature to the face recognition platform.

[0065] In order to effectively save server resources and improve the efficiency and accuracy of face recognition, the present application provides an embodiment of a face recognition method, the execution subject can be a face recognition platform deployed in the cloud, see Figure 3 , the face recognition method specifically includes the following contents:

[0066] Step S301: receiving a facial image, a first facial feature and a feature matching result sent by at least one face capture camera, wherein the first facial feature is extracted according to a first image recognition algorithm after the face capture camera captures the facial image, and the feature matching result is obtained by the face capture camera storing the facial feature in a local feature queue and then performing feature matching with a local target feature library, and the local target feature library of the face capture camera stores target facial feature values ​​sent by a face recognition platform.

[0067] Step S302: Determine whether the feature matching result is passed. If so, extract the second facial feature corresponding to the facial image according to the second image recognition algorithm, and perform feature matching on the second facial feature with the local target feature library to obtain the facial recognition result. Otherwise, send the first facial feature to the local feature queue of the corresponding facial capture camera.

[0068] From the above description, it can be seen that the face recognition method provided in the embodiment of the present application can form a distributed network established by multiple face capture cameras by utilizing the face search support function of the face capture camera itself, and utilize the numerous face capture cameras on the distributed network to realize their respective face search functions. Especially in the case of massive data, it can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0069] In order to improve the accuracy of face image recognition, in one embodiment of the face recognition method of the present application, see Figure 4 Step S202 may also specifically include the following contents:

[0070] Step S401: extracting features from the face image according to the target person information to be confirmed in the matching result, the face image corresponding to the target person information to be confirmed, and a second image recognition algorithm to obtain a second face feature.

[0071] Step S402: determine whether the second facial feature matches the local target feature library. If so, obtain the corresponding face recognition result. Otherwise, send the first facial feature to the local feature queue of the corresponding face capture camera.

[0072] In order to effectively save server resources and improve the efficiency and accuracy of face recognition, the present application provides an embodiment of a face recognition device for implementing all or part of the face recognition method. Figure 5 , the face recognition device specifically includes the following contents:

[0073] The face capture camera image analysis module 10 is used to capture a face image and extract a first face feature according to a first image recognition algorithm, and store the face feature in a local feature queue.

[0074] The face capture camera information sending module 20 is used to perform feature matching on the face features in the local feature queue and the local target feature library, and send the feature matching result, the face image and the first face feature to the face recognition platform, wherein the local target feature library stores the target face feature value sent by the face recognition platform, so that the face recognition platform extracts the second face feature corresponding to the face image according to the second image recognition algorithm when the feature matching result is passed, and performs feature matching on the second face feature and the local target feature library to obtain the face recognition result, and the face recognition platform sends the first face feature to the local feature queue of the corresponding face capture camera when the feature matching result is failed.

[0075] From the above description, it can be seen that the face recognition device provided in the embodiment of the present application can form a distributed network established by multiple face capture cameras by utilizing the face search support function of the face capture camera itself, and utilize the numerous face capture cameras on the distributed network to realize their respective face search functions. Especially in the case of massive data, it can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0076] In order to ensure the reliability of face image recognition and avoid data loss or recognition delay caused by the performance problem of the face capture camera itself, in one embodiment of the face recognition device of the present application, see Figure 6 , and specifically includes the following contents:

[0077] The face capture camera performance monitoring unit 51 is used to determine whether the local feature queue exceeds the feature quantity threshold or whether the current CPU usage exceeds the operation threshold.

[0078] The performance abnormality sending unit 52 is used to terminate the feature matching operation and send the face image and the first face feature to the face recognition platform if the local feature queue exceeds the feature quantity threshold or the current CPU usage exceeds the operation threshold.

[0079] In order to effectively save server resources and improve the efficiency and accuracy of face recognition, the present application provides an embodiment of a face recognition device for implementing all or part of the face recognition method. Figure 7 , the face recognition device specifically includes the following contents:

[0080] The face recognition platform image analysis module 30 is used to receive a face image, a first face feature and a feature matching result sent by at least one face capture camera, wherein the first face feature is extracted according to a first image recognition algorithm after the face capture camera captures the face image, and the feature matching result is obtained by the face capture camera storing the face feature in a local feature queue and then performing feature matching with the face features in the local feature queue and a local target feature library, and the local target feature library of the face capture camera stores the target face feature values ​​sent by the face recognition platform.

[0081] The face recognition platform information processing module 40 is used to determine whether the feature matching result is passed. If so, the second face feature corresponding to the face image is extracted according to the second image recognition algorithm, and the second face feature is feature matched with the local target feature library to obtain the face recognition result. Otherwise, the first face feature is sent to the local feature queue of the corresponding face capture camera.

[0082] From the above description, it can be seen that the face recognition device provided in the embodiment of the present application can form a distributed network established by multiple face capture cameras by utilizing the face search support function of the face capture camera itself, and utilize the numerous face capture cameras on the distributed network to realize their respective face search functions. Especially in the case of massive data, it can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0083] In order to illustrate the present solution, the present application also provides a face recognition system that uses the above-mentioned face recognition device to implement the face recognition method. Figure 8 , specifically including the following contents: a face capture camera and a face recognition platform communicatively connected to the face capture camera.

[0084] The face capture camera comprises:

[0085] The face capture camera image analysis module 10 is used to capture a face image and extract a first face feature according to a first image recognition algorithm, and store the face feature in a local feature queue.

[0086] The face capture camera information sending module 20 is used to perform feature matching on the face features in the local feature queue and the local target feature library, and send the feature matching results, the face image and the first face feature to the face recognition platform, wherein the local target feature library stores the target face feature values ​​sent by the face recognition platform.

[0087] The face recognition platform includes:

[0088] The face recognition platform image analysis module 30 is used to receive a face image, a first face feature and a feature matching result sent by at least one face capture camera.

[0089] The face recognition platform information processing module 40 is used to determine whether the feature matching result is passed. If so, the second face feature corresponding to the face image is extracted according to the second image recognition algorithm, and the second face feature is feature matched with the local target feature library to obtain the face recognition result. Otherwise, the first face feature is sent to the local feature queue of the corresponding face capture camera.

[0090] From the above description, it can be seen that the face recognition system provided in the embodiment of the present application can form a distributed network established by multiple face capture cameras by utilizing the face search support function of the face capture camera itself, and utilize the numerous face capture cameras on the distributed network to realize their respective face search functions. Especially in the case of massive data, it can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0091] In order to further illustrate the present solution, the present application also provides a specific application example of a face recognition system using the above-mentioned face recognition device to implement the face recognition method. Fig. 9 , including terminal face recognition and cloud face recognition. Among them: terminal face recognition includes several face capture cameras 1 and the face search function they provide; cloud face recognition is a face recognition platform 2 and the face search function it provides.

[0092] See also Fig.10 The following is the process flow of face search:

[0093] Step S601: The user passes through the shooting range of the face capture camera in a natural manner.

[0094] Step S602: The image acquisition module of the face capture camera acquires the image, obtains the target image, and continues the subsequent processing flow.

[0095] Step S603: the portrait detection module of the face capture camera detects the target image, determines that the image contains portrait information, and continues the subsequent processing flow.

[0096] Step S604: the face capture module of the face capture camera captures the portrait information to obtain a relatively clear initial target person's face image.

[0097] Step S605: The feature extraction module of the face capture camera performs feature extraction on the obtained initial target person's face image to obtain a face feature value.

[0098] Step S606: putting the facial feature value into the feature queue.

[0099] Step S607: The face search module of the face capture camera determines whether there is data in the feature queue. If there is, a face feature value is read from the feature queue in a first-in-first-out manner, and a face search is performed in the terminal face feature library in the camera according to the corresponding algorithm, and the result is sent to the face recognition platform.

[0100] The result data sent is divided into two categories according to different identifiers:

[0101] Category 1: marked as "search successful", that is, the target person information to be confirmed that meets the threshold can be searched in the face capture camera. The target person information to be confirmed and the initial target face image are sent to the face recognition platform.

[0102] Category 2: marked as "search failure", that is, there is no result that meets the threshold in the face capture camera. Then the current face feature value and the initial target face image are sent to the face recognition platform (if the current face feature value is sent from other face capture cameras, there is no need to send the initial target face image)

[0103] The above steps are executed repeatedly until all data in the feature queue are processed.

[0104] Step S608: The result sent by the face capture camera is placed in the message center.

[0105] Step S609: The face search interface of the face recognition platform subscribes to data from the message center.

[0106] Step S610: The system analyzes and judges the read data. If the data is marked as "search successful", the subsequent cloud face search continues; otherwise, jump to step S612.

[0107] Step S611: The cloud-based face search logic of the face recognition platform reads the initial target face image from the data and uses other high-precision face recognition algorithms to extract face feature values. At the same time, based on the target person information to be confirmed in the data, the feature values ​​of the corresponding high-precision algorithm are queried from the face feature library in the cloud. Then, a high-precision algorithm is used to perform a 1:N face search in a small range. Finally, the results are sorted, and the result with the highest accuracy is the final target person information. The entire face search process ends.

[0108] Step S612: If the data is marked as "search failed", it means that the target person who meets the threshold cannot be found in the current camera, then the facial feature value to be searched is sent to other face capture cameras on the network for face search according to the rules. The process jumps to step S606.

[0109] From the above content, it can be seen that the face recognition system of the present application utilizes the face capture camera to support the face search function, constitutes a large number of shared face capture cameras on the network, and establishes a distributed face recognition system on the network, which has the following effects and advantages:

[0110] 1. Save resources. Using the numerous face capture cameras on the Internet to realize the face search function can greatly save server resources, especially in the case of massive data.

[0111] 2. High security. The system combines terminal face recognition and cloud face recognition, and applies different algorithms to enhance the security of face recognition. Only when both the terminal and the cloud recognize the target user and verify it through different algorithms can the user's identity information be finally determined.

[0112] From the hardware level, in order to effectively save server resources and improve the efficiency and accuracy of face recognition, the present application provides an embodiment of an electronic device for implementing all or part of the content of the face recognition method, and the electronic device specifically includes the following content:

[0113] Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the face recognition device and related devices such as the core business system, user terminal and related database; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the face recognition method and the embodiment of the face recognition device in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0114] It is understandable that the user terminal may 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 may include smart glasses, a smart watch, a smart bracelet, etc.

[0115] In practical applications, part of the face recognition method can be executed on the electronic device side as described above, or all operations can 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.

[0116] 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.

[0117] Fig.11 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.11 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.11 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0118] In one embodiment, the face recognition method function may be integrated into the central processing unit 9100. The central processing unit 9100 may be configured to perform the following control:

[0119] Step S101: capturing a facial image and extracting a first facial feature according to a first image recognition algorithm, and storing the facial feature in a local feature queue.

[0120] Step S102: Feature matching is performed on the facial features in the local feature queue and the local target feature library, and the feature matching result, the facial image and the first facial feature are sent to the facial recognition platform, wherein the local target feature library stores the target facial feature value sent by the facial recognition platform, so that the facial recognition platform extracts the second facial feature corresponding to the facial image according to the second image recognition algorithm when the feature matching result is passed, and feature matches the second facial feature with the local target feature library to obtain the facial recognition result, and the facial recognition platform sends the first facial feature to the local feature queue of the corresponding facial capture camera when the feature matching result is failed.

[0121] From the above description, it can be seen that the electronic device provided in the embodiment of the present application, by utilizing the face capture camera's own function of supporting face search, forms a distributed network established by multiple face capture cameras, and utilizes the numerous face capture cameras on the distributed network to realize their respective face search functions. Especially in the case of massive data, it can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0122] In another embodiment, the face recognition device can be configured separately from the central processing unit 9100. For example, the face recognition device can be configured as a chip connected to the central processing unit 9100, and the face recognition method function is implemented under the control of the central processing unit.

[0123] like Fig.11 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.11 In addition, the electronic device 9600 may also include Fig.11 For components not shown, reference may be made to the prior art.

[0124] like Fig.11 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.

[0125] 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.

[0126] 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.

[0127] 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.

[0128] 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.).

[0129] 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.

[0130] 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.

[0131] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the face recognition method in the above embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all steps of the face recognition method in the above embodiment, where the execution subject is a server or a client, are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0132] Step S101: capturing a facial image and extracting a first facial feature according to a first image recognition algorithm, and storing the facial feature in a local feature queue.

[0133] Step S102: Feature matching is performed on the facial features in the local feature queue and the local target feature library, and the feature matching result, the facial image and the first facial feature are sent to the facial recognition platform, wherein the local target feature library stores the target facial feature value sent by the facial recognition platform, so that the facial recognition platform extracts the second facial feature corresponding to the facial image according to the second image recognition algorithm when the feature matching result is passed, and feature matches the second facial feature with the local target feature library to obtain the facial recognition result, and the facial recognition platform sends the first facial feature to the local feature queue of the corresponding facial capture camera when the feature matching result is failed.

[0134] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application, by utilizing the face capture camera's own function of supporting face search, forms a distributed network established by multiple face capture cameras, and utilizes the numerous face capture cameras on the distributed network to realize their respective face search functions, especially in the case of massive data, which can greatly save server resources. At the same time, through the combination of terminal face recognition and cloud face recognition, different face recognition algorithms are applied to enhance the security and accuracy of face recognition. That is, only when both the terminal and the cloud recognize the target face and are verified by different algorithms can the identity information of the target be finally determined, thereby effectively saving server resources and improving face recognition efficiency and accuracy.

[0135] It will be appreciated by those skilled in the art that 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. 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 code.

[0136] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0137] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0138] 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 The steps for the functions specified in one or more boxes.

[0139] 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 face recognition method, characterized in that: The method comprises: Capturing a facial image and extracting a first facial feature according to a first image recognition algorithm, and storing the facial feature in a local feature queue; The facial features in the local feature queue are feature matched with the local target feature library, and the feature matching result, the facial image and the first facial feature are sent to the facial recognition platform, wherein the local target feature library stores the target facial feature value sent by the facial recognition platform, so that the facial recognition platform extracts the second facial feature corresponding to the facial image according to the second image recognition algorithm when the feature matching result is passed, and feature matches the second facial feature with the local target feature library to obtain the facial recognition result, and when the feature matching result is failed, the facial recognition platform sends the first facial feature to the local feature queue of other facial capture cameras, and continues to search for faces in other facial capture cameras; the feature value data in the local feature queue comes from the following two parts: the first facial feature obtained by feature extraction and the feature value sent from the facial recognition platform.

2. The face recognition method according to claim 1, characterized in that: After storing the facial features in the local feature queue, the method further includes: Determine whether the local feature queue exceeds a feature quantity threshold or whether the current CPU usage exceeds an operation threshold; If so, the feature matching operation is terminated and the facial image and the first facial feature are sent to the facial recognition platform.

3. A face recognition method, characterized in that: The method comprises: Receive a face image, a first face feature, and a feature matching result sent by at least one face capture camera, wherein the first face feature is extracted according to a first image recognition algorithm after the face capture camera captures the face image, and the feature matching result is obtained by the face capture camera storing the face feature in a local feature queue and then performing feature matching on the face features in the local feature queue with a local target feature library, and the local target feature library of the face capture camera stores target face feature values ​​sent by a face recognition platform; Determine whether the feature matching result is passed. If so, extract the second facial feature corresponding to the facial image according to the second image recognition algorithm, and perform feature matching on the second facial feature and the local target feature library to obtain the facial recognition result. Otherwise, send the first facial feature to the local feature queue of other facial capture cameras, and continue to search for faces in other facial capture cameras. The feature value data in the local feature queue comes from the following two parts: the first facial feature obtained by feature extraction and the feature value sent from the facial recognition platform.

4. The face recognition method according to claim 3, characterized in that: The extracting a second facial feature corresponding to the facial image according to a second image recognition algorithm, and performing feature matching on the second facial feature with a local target feature library to obtain a facial recognition result includes: Extracting features from the face image according to the target person information to be confirmed in the matching result, the face image corresponding to the target person information to be confirmed, and a second image recognition algorithm to obtain a second face feature; Determine whether the second facial feature matches the local target feature library. If so, obtain the corresponding face recognition result. Otherwise, send the first facial feature to the local feature queue of other face capture cameras, and continue to search for faces in other face capture cameras.

5. A face recognition device, characterized in that: include: A face capture camera image analysis module, used to capture a face image and extract a first face feature according to a first image recognition algorithm, and store the face feature in a local feature queue; The face capture camera information uploading module is used to perform feature matching between the face features in the local feature queue and the local target feature library, and send the feature matching result, the face image and the first face feature to the face recognition platform, wherein the local target feature library stores the target face feature value sent by the face recognition platform, so that the face recognition platform extracts the second face feature corresponding to the face image according to the second image recognition algorithm when the feature matching result is passed, and performs feature matching between the second face feature and the local target feature library to obtain the face recognition result, and when the feature matching result is not passed, the face recognition platform sends the first face feature to the local feature queue of other face capture cameras, and continues to search for faces in other face capture cameras; the feature value data in the local feature queue comes from the following two parts: the first face feature obtained by feature extraction and the feature value sent from the face recognition platform.

6. The face recognition device according to claim 5, characterized in that: Also includes: A face capture camera performance monitoring unit, used to determine whether the local feature queue exceeds a feature quantity threshold or whether the current CPU usage exceeds an operating threshold; A performance abnormality sending unit is used to terminate the feature matching operation and send the face image and the first face feature to the face recognition platform if the local feature queue exceeds the feature quantity threshold or the current CPU usage exceeds the operation threshold.

7. A face recognition device, characterized in that: include: A face recognition platform image analysis module, used for receiving a face image, a first face feature and a feature matching result sent by at least one face capture camera, wherein the first face feature is extracted according to a first image recognition algorithm after the face capture camera captures the face image, and the feature matching result is obtained by the face capture camera storing the face feature in a local feature queue and then performing feature matching on the face features in the local feature queue with a local target feature library, and the local target feature library of the face capture camera stores target face feature values ​​sent by the face recognition platform; The face recognition platform information processing module is used to determine whether the feature matching result is passed. If so, the second face feature corresponding to the face image is extracted according to the second image recognition algorithm, and the second face feature is feature matched with the local target feature library to obtain the face recognition result. Otherwise, the first face feature is sent to the local feature queue of other face capture cameras, and the face search is continued in other face capture cameras; the feature value data in the local feature queue comes from the following two parts: the first face feature obtained by feature extraction and the feature value sent from the face recognition platform.

8. A face recognition system, characterized in that: include: A face capture camera and a face recognition platform communicatively connected to the face capture camera; The face capture camera comprises: A face capture camera image analysis module, used to capture a face image and extract a first face feature according to a first image recognition algorithm, and store the face feature in a local feature queue; A face capture camera information uploading module, used for matching the face features in the local feature queue with the local target feature library, and sending the feature matching result, the face image and the first face feature to the face recognition platform, wherein the local target feature library stores the target face feature value sent by the face recognition platform; The face recognition platform includes: A face recognition platform image analysis module, used for receiving a face image, a first face feature and a feature matching result sent by at least one face capture camera; The face recognition platform information processing module is used to determine whether the feature matching result is passed. If so, the second face feature corresponding to the face image is extracted according to the second image recognition algorithm, and the second face feature is feature matched with the local target feature library to obtain the face recognition result. Otherwise, the first face feature is sent to the local feature queue of other face capture cameras, and the face search is continued in other face capture cameras; the feature value data in the local feature queue comes from the following two parts: the first face feature obtained by feature extraction and the feature value sent from the face recognition platform.

9. 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 program, the steps of the face recognition method according to any one of claims 1 to 4 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the face recognition method according to any one of claims 1 to 4 are implemented.

Images