WO2021047664A1

WO2021047664A1 - Biometric feature recognition method and related device

Info

Publication number: WO2021047664A1
Application number: PCT/CN2020/114975
Authority: WO
Inventors: 曹俊; 刘芬
Original assignee: 华为技术有限公司
Priority date: 2019-09-12
Filing date: 2020-09-14
Publication date: 2021-03-18
Also published as: CN112487082A; CN112487082B

Abstract

A biometric feature recognition method and a related device. The method comprises: a first edge node receiving broadcast information that is sent by a second edge node and includes a first biometric feature; the first edge node determining that the probability value of the first biometric feature appearing in a management range of the first edge node is greater than or equal to a preset threshold value, and the first edge node storing the first biometric feature in a first edge biometric feature database; and the first edge node recognizing, by using the first edge biometric feature database, biometric features collected from within the management range of the first edge node. By means of the method, the response speed of biometric feature recognition can be improved, thereby ensuring the real-time performance and the stability of biometric feature recognition.

Description

Method and related equipment for biometric identification

Technical field

This application relates to the field of biometric identification technology, and in particular to a method and related equipment for biometric identification.

Background technique

Biometric recognition, such as face recognition, iris recognition, gait recognition, etc., is the use of human physiological or behavioral characteristics to identify personal identity. Among them, face recognition is an important biometric recognition technology, which is a general term for technologies that automatically determine the similarity of two face images through a computer. At present, face recognition is widely used in various industries and scenarios such as security, retail, finance, and office. To realize face recognition, it is first necessary to build a face database, which is completed through the steps of camera shooting, capturing face images, face comparison, and identity recognition. It has high requirements for real-time and stability, and usually requires less than 1 second for response time.

However, the scale of the biometric database (such as the face database) is generally large and requires large storage resources. In addition, the amount of calculation for biometric comparison requires large computing resources. In order to meet the storage resources and computing resources required for biometric identification, the biometric database is generally established in a cloud environment, and biometric identification is also performed in the cloud environment. However, the cloud environment is far away from terminal devices (such as cameras, sensors, etc.), the time delay for transmitting biometric information to the cloud environment is relatively large, and it is easily affected by factors such as network fluctuations, and cannot meet the requirements of real-time and stability.

Therefore, how to reduce the time delay of biometric identification and ensure the requirements of real-time and stability of biometric identification are technical problems to be solved urgently.

Summary of the invention

This application provides a biometric identification method and related equipment, which can ensure the real-time and stability requirements of biometric identification, and improve the response speed of biometric identification.

In a first aspect, a method for biometric identification is provided, including: a first edge node receives broadcast information sent by a second edge node, the broadcast information includes the first biometric feature, and the first edge node stores the first edge biometric A feature library, the first edge biometric feature library includes at least one biometric feature; the first edge node determines that the probability value of the first biometric feature appearing in the management range of the first edge node is greater than or equal to a preset threshold, the The first edge node stores the first biometric feature in the first edge biometric database; the first edge node uses the first edge biometric database to analyze the biometric features collected within the management range of the first edge node Identify it.

In the embodiment of the present application, the first edge node judges the received first biological feature broadcast by the second edge node, and the probability value of the first biological feature appearing in the management range of the first edge node is greater than or If it is equal to the preset threshold, store it in the first biometric database, and use the first biometric database to perform biometric identification, which can improve the response speed of biometric identification and ensure the real-time and stability of biometric identification.

With reference to the first aspect, in a possible implementation of the first aspect, the first edge node queries a cloud manager for social relationship information corresponding to the first biometric feature, and the cloud manager includes a social relationship database, so The social relationship database includes social relationship information corresponding to biological characteristics; the first edge node calculates the degree of association between the first biological characteristics and the first edge node according to the social relationship information corresponding to the first biological characteristics, The degree of association is used to indicate the probability that the first biological feature appears within the management range of the first edge node; the first edge node determines that the value of the degree of association is greater than or equal to the preset threshold.

In the embodiment of the present application, the first edge node uses the social relationship information corresponding to the first biometric feature to calculate the degree of association between the first edge node and the first biometric feature, and determines whether to store the relationship based on the relationship between the degree of association and a preset threshold. The first biological feature can ensure that the stored first biological feature has a high degree of association with the first edge node, that is, the first biological feature has a greater probability of appearing in the management range of the first edge node, which can improve the first biological feature. The recognition speed of features ensures the real-time and stability of recognition.

With reference to the first aspect, in a possible implementation of the first aspect, the first edge node reads the biological characteristics in the first edge biometric database, and queries the cloud manager for the first The social relationship intimacy between the biological feature and the biological feature in the first edge biological feature database; the first edge node queries the cloud manager for personal tag information and activity track information corresponding to the first biological feature.

In the embodiment of the present application, the first edge node queries the social relationship intimacy between the first biometric feature and the biometric feature in the first edge biometric database, and the personal tag information and activity track information corresponding to the first biometric feature. Obtaining the social relationship information corresponding to the first biometric feature ensures the comprehensiveness of the obtained social relationship information, and ensures the reliability of the calculated association degree.

With reference to the first aspect, in a possible implementation of the first aspect, the broadcast information further includes geographic location information of the second edge node and label information of the second edge node; the first edge node According to the label information of the first edge node and the label information of the second edge node, the first label similarity is calculated, and the label information of the first edge node is used to indicate the type of the first edge node, so The label information of the second edge node is used to indicate the type of the second edge node; the first edge node calculates the The distance between the first edge node and the second edge node is calculated based on the distance and the activity track information; the first edge node is based on the label information of the first edge node and the first edge node. The personal tag information corresponding to a biological feature is calculated, and the second tag similarity is calculated; the first edge node is based on the social relationship intimacy, the first tag similarity, the second tag similarity, and the geographic location Any one or more of the similarities is calculated, and the degree of association between the first biological feature and the first edge node is calculated.

In the embodiment of this application, the first edge node calculates the first edge similarity, the second tag similarity, the geographic location similarity, and the social relationship intimacy first, and then further calculates the first edge based on one or more of them. The degree of association between the node and the first biological feature can ensure that the value of the degree of association can be calculated freely and flexibly, thereby improving the real-time and stability of biometric identification.

With reference to the first aspect, in a possible implementation of the first aspect, when the storage resource of the first edge node is less than a preset storage threshold, the first edge node deletes the biometric feature according to a preset rule; The preset rules include: deleting the biometrics whose appearance times are less than the first threshold; or, deleting the biometrics whose similarity with the label information of the first edge node is less than the second threshold; or, if the biometrics appearing last The biological features whose time distance exceeds the third threshold from the current time are deleted.

In the embodiment of the present application, when the storage resources are insufficient, the first edge node deletes some biometrics according to preset rules, which can ensure that the first edge node can reserve enough storage resources to store the biometrics broadcast by other edge nodes. The real-time and stability of biometrics.

With reference to the first aspect, in a possible implementation of the first aspect, the first edge node queries the cloud manager for address information of a third edge node whose distance from the first edge node is less than a preset distance, The cloud manager includes an edge node information database, and the edge node information database includes geographic location information and address information of the edge node; the first edge node sends the information to the first edge node according to the address information of the third edge node. The three-edge node sends second broadcast information, where the second broadcast information includes the second biometric feature.

In the embodiment of this application, the first edge node not only receives the biological characteristics broadcast by other neighboring edge nodes, but the first edge node can also obtain the address information of other neighboring edge nodes through query, and then broadcast the collected data to other edge nodes. The second biometric feature ensures that other edge nodes can judge and store the second biometric feature, thereby improving the real-time and stability of biometric identification.

In a second aspect, the present application provides an edge node, including: an edge biometrics database for storing at least one biometric; a biometric extraction unit for receiving broadcast information from a second edge node and extracting the broadcast information including A first biological feature; a biological feature management unit for determining that the probability value of the first biological feature appearing in the range managed by the edge node is greater than or equal to a preset threshold, and storing the first biological feature in the Edge biometric database; a biometric identification unit for identifying biometrics collected within the management range of the edge node by using the edge biometric database.

With reference to the second aspect, in a possible implementation of the second aspect, the biometrics management unit is further configured to query the cloud manager for social relationship information corresponding to the first biometric, and the cloud management The device includes a social relationship database, which includes social relationship information corresponding to biological characteristics; according to the social relationship information corresponding to the first biological characteristics, the degree of association between the first biological characteristics and the edge node is calculated, so The degree of association is used to indicate the probability that the first biological feature appears within the management range of the edge node; it is determined that the value of the degree of association is greater than or equal to the preset threshold.

With reference to the second aspect, in a possible implementation of the second aspect, the biometrics management unit is further configured to: read the biometrics in the edge biometric database, and query the cloud manager for all the biometrics. The social relationship intimacy between the first biological feature and the biological feature in the edge biological feature database; and query the cloud manager for personal tag information and activity track information corresponding to the first biological feature.

With reference to the second aspect, in a possible implementation of the second aspect, the broadcast information further includes geographic location information of the second edge node and label information of the second edge node; the biometric management unit , Is also used to: calculate a first label similarity according to the label information of the edge node and the label information of the second edge node, the label information of the edge node is used to indicate the type of the edge node, the The label information of the second edge node is used to indicate the type of the second edge node; the edge node and the second edge are calculated according to the geographic location information of the edge node and the geographic location information of the second edge node The distance of the node is calculated based on the distance and the activity trajectory information; the second tag similarity is calculated based on the tag information of the edge node and the personal tag information corresponding to the first biological feature; Any one or more of the social relationship intimacy, the first tag similarity, the second tag similarity, and the geographic location similarity, and the association between the first biological feature and the edge node is calculated degree.

With reference to the second aspect, in a possible implementation of the second aspect, the biometrics management unit is further configured to: when the storage resource of the edge node is less than a preset storage threshold, delete the biometrics according to a preset rule The preset rule includes: deleting the biometrics whose appearance times are less than the first threshold; or, deleting the biometrics whose similarity with the label information of the edge node is less than the second threshold; or, The biological features whose last appearance time exceeds the third threshold from the current time are deleted.

With reference to the second aspect, in a possible implementation of the second aspect, the biometric management unit is further configured to query the cloud manager for the address of a third edge node whose distance from the edge node is less than a preset distance Information, the cloud manager includes an edge node information database, the edge node information database includes geographic location information and address information of the edge node; the biometric management unit is further configured to Information, sending second broadcast information to the third edge node, where the second broadcast information includes a second biological feature.

In a third aspect, the present application provides a computing device cluster. The computing device cluster includes at least one computing device, each computing device includes a processor and a memory, and the processor of the at least one computing device is used to call the at least one computing device. The program code in the memory of the computing device executes the above-mentioned first aspect and the method in combination with any one of the above-mentioned first aspect.

In a fourth aspect, the present application provides a computer storage medium, the computer storage medium stores a computer program, and when the computer program is executed by a processor, the above first aspect is implemented and any one of the above first aspect is implemented. The flow of the method provided by the method.

In a fifth aspect, the present application provides a computer program, the computer program includes instructions, when the computer program is executed by a computer, the computer can execute the above-mentioned first aspect and any one of the above-mentioned implementation methods of the first aspect. The flow of the provided method.

Description of the drawings

FIG. 1 is a schematic diagram of a flow of biometric identification according to an embodiment of this application;

2 is a schematic structural diagram of a face recognition system provided by an embodiment of the application;

FIG. 3 is a schematic structural diagram of a biometric identification system provided by an embodiment of this application;

FIG. 4 is a schematic diagram of a system architecture provided by an embodiment of the application;

FIG. 5 is a schematic structural diagram of another biometric identification system provided by an embodiment of this application;

FIG. 6 is a schematic flowchart of a method for biometric identification according to an embodiment of this application;

FIG. 7 is a schematic structural diagram of a computing device provided by an embodiment of this application;

FIG. 8 is a schematic structural diagram of a computing device cluster provided by an embodiment of this application.

detailed description

The following describes the technical solutions in the embodiments of the present application clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

First of all, some terms and related technologies involved in this application will be explained in conjunction with the accompanying drawings, so as to facilitate the understanding of those skilled in the art.

Biometric identification technology refers to the close combination of computers and high-tech means such as optics, acoustics, biosensors, and biostatistics principles, and the use of inherent physiological and behavioral characteristics of the human body to identify personal identity. It has been used in biometrics. Biological characteristics include hand shape, fingerprint, face shape, iris, retina, pulse, auricle, etc.

Refer to Figure 1, which is a schematic diagram of a flow of biometric identification. As shown in Figure 1, the information collection module 110 first needs to perform information collection, for example, using a camera or infrared camera to collect a face image, and then the preprocessing module 120 performs preprocessing. The data containing biometric information is processed to determine the area where the biometrics are located. Then the feature processing module 130 performs feature processing, and further processes the pre-processed information, that is, converts the biometric information into a string of "digital codes" that characterize its characteristics, and stores the finally obtained biometrics in the biometric database 140, In order to facilitate subsequent comparison and identification. After the establishment of the biometric database 140 is completed, the biometric database 140 can be used to identify the biometric to be recognized. Specifically, the information collection module 110 collects data including biometric information to be identified, the preprocessing module 120 preprocesses the data, and then the feature processing module 130 performs feature processing to obtain the biometric features to be identified, and then the feature comparison module 150 will The obtained biological characteristic is compared with the biological characteristic stored in the biological characteristic database 140 to identify the identity of the biological characteristic.

It should be noted that to complete biometric identification, it is necessary to construct a biometric database and conduct biometric comparison in advance. The scale of the biometric database is relatively large, ranging from one million to one hundred million, which requires a large storage space for storage; in addition, because the biometric database is large, the amount of calculation is also large when comparing biometrics. That is, the required computing resources are relatively large. Cloud environment refers to the central computing equipment cluster owned by cloud service providers to provide computing, storage, and communication resources. It has large storage resources and computing resources to meet the requirements of biometric identification. Therefore, the establishment of biometric databases and The comparison of biological characteristics is carried out in the cloud environment.

It should be understood that although the cloud environment can meet the storage and calculation requirements of biometrics, the cloud environment is far away from terminal devices (such as cameras, sensors, etc.). When performing biometrics, the biometrics need to be uploaded to the cloud environment. After the recognition is completed, the cloud environment will return the recognition result. As shown in FIG. 2, in face recognition, the camera 210 collects a face image and sends it to the edge node 220. The face feature extraction module 221 in the edge node 220 processes the received face image to extract the face. Then upload the facial features to the central computing device 230 in the cloud environment. The central computing device 230 compares the received facial features with the facial features in the face database 231 one by one. If there is a facial feature matching the received facial feature in 231, the identity information corresponding to the facial feature is returned to the edge node 220; if there is no facial feature matching the received facial feature in the face database 231 If the matching facial feature is matched, then a message that the facial feature does not exist is returned to the edge node 220. Although face recognition can be completed by the above method, it takes a long time to transmit facial features to the cloud environment and return the recognition result from the cloud environment, and the time delay is large, and it is easily affected by factors such as network fluctuations during the transmission process.

In order to reduce the time delay of biometric identification, the edge nodes in the edge environment (ie, the edge computing device cluster that is close to the terminal device in the geographical position and used to provide computing, storage, and communication resources) can be used to share part of the cloud environment. The task and function of the central computing device in the central computing device is to build an edge biometric database in the edge node. The edge biometric database contains some of the biometric data in the central biometric database (the biometric database stored in the cloud environment) , And use the edge biometric database to complete biometric identification. As shown in FIG. 3, the edge node 320 receives the biometric data collected by the biometric data collection device 310, the biometric extraction module 321 in the edge node 320 processes the data, extracts the biometrics, and then compares the biometrics with the biometrics. The biological characteristics in the edge biological characteristic database 322 are compared one by one to obtain the result of the biological characteristic recognition.

The above-mentioned biometric identification methods can enable the biometric identification process to be completed locally, but due to the limited storage and computing resources of edge nodes, it can only support small-scale edge biometric databases, which cannot meet large-scale biometric identification scenarios. Demand. In addition, if no matching biometric is found in the edge database, it still needs to be uploaded to the cloud environment, and the cloud environment completes the biometric identification process.

In summary, the present application provides a method and related equipment for biometric identification, which can improve the real-time performance and stability of biometric identification, and improve the response speed of biometric identification.

The technical solutions of the embodiments of the present application can be applied to various biometric recognition scenarios, including but not limited to face recognition, fingerprint recognition, iris recognition, and the like.

In a specific embodiment, as shown in FIG. 4, the biometric identification system can be deployed in a cloud environment and an edge environment, specifically one or more computing devices (such as a central server) in the cloud environment and one of the edge environments. Or on multiple computing devices (edge computing devices), the edge computing device may be a server. The original data collection equipment collects biometric data required for biometric identification, including but not limited to cameras, infrared cameras, etc. In addition, the biometric identification system includes multiple parts (for example, multiple subsystems, each of which includes multiple units), and each part can be deployed in different environments in a distributed manner. For example, a part of the biometric recognition system can be separately deployed in three of the cloud environment, the edge environment, the original data collection device, or any two of them.

The biometric identification system is used to perform biometric identification based on the biometric data collected by the original data collection device. The internal units of the biometric identification system can be divided into multiple ways, which is not limited in this application. FIG. 5 is an exemplary division method. As shown in FIG. 5, the biometric identification system 500 includes a plurality of edge nodes 510 and a central node 520. The functions of each device and its functional units are briefly described below.

The edge node 510 shown is used to receive biometric data collected by at least one original data collection device. The edge node 510 includes multiple functional units. Among them, the edge biometric database 511 is used to store the biometrics cached by the edge node; the biometric extraction unit 512 is used to extract biometrics in the biometric data; the biometric identification unit 513 , Used to compare the biological characteristics obtained by the biological characteristic extraction unit 512 with the biological characteristics stored in the edge biological characteristic database 511 to complete the biological characteristic identification; the biological characteristic management unit 514 is used to compare the edge biological characteristic database 511 Manage the stored biological characteristics, such as broadcasting biological characteristics to neighboring edge nodes, broadcasting the node information of the edge node (such as geographic location information, label information, etc.), and whether to broadcast to other neighboring edge nodes received Biometrics are stored.

Optionally, the edge node 510 further includes a service processing unit 515, configured to perform corresponding service processing (for example, a stranger alarm is recognized, etc.) after the biometric recognition unit 513 completes biometric recognition.

The central node 520 shown is used to communicate with the edge node 510 and assist the edge node 510 to complete biometric identification. The central node 520 includes multiple functional units. Among them, the central biometric database 521 is used to store all facial features in an area, that is, the central biometric database 521 stores all the biometrics stored in the edge biometric database 511; The relationship database 522 is used to store the social relationship information corresponding to each biological feature, such as other biological characteristics, personal tags, activity trajectories, etc. associated with each biological feature; the edge node location information management unit 523 is used to store and manage all edges The geographic location information (such as latitude and longitude, etc.) and address information (such as IP address, etc.) corresponding to the edge node label of the node.

In this application, the biometric identification system 500 may be a software system. The various parts and functional units included in it are deployed on hardware devices in a flexible manner. As shown in FIG. 4, the entire system can be deployed in two or three distributed ways. One or more computing devices in one environment.

Please refer to FIG. 6, which is a schematic flowchart of a method for biometric identification according to an embodiment of this application. As shown in Figure 6, the method includes but is not limited to the following steps:

S601: The first edge node receives broadcast information sent by the second edge node.

Specifically, the first edge node and the second edge node may be computing devices in an edge environment, and the broadcast information sent by the first edge node includes a first biometric feature. The biometric feature may be a face feature, a fingerprint feature, or an iris feature. Wait.

Further, the first edge node and the second edge node are each bound with one or more cameras, and the biometric data (for example, a face image) can be collected by using the cameras. In addition, they each include an edge biometric database, which may be the edge biometric database 511 shown in FIG. 5, the edge biometric database stores a plurality of biometrics, and after the biometric data is collected, The biological feature extraction unit 512 can be used to perform feature extraction on the biological feature data to obtain the corresponding biological features, and then the biological feature recognition unit 513 compares it with the biological features in the edge biological feature database 511 to complete the extraction. Recognition of biological characteristics, and corresponding processing according to the recognition results.

It can be understood that if the edge biometric database can be used to complete the identification of the biometrics to be identified, it is not necessary to send the biometrics to be identified to the central node 520 in the cloud environment, which reduces the time delay of the biometric identification process and improves the biometric identification. speed.

The second edge node uses the second edge biometric database to perform biometric identification after collecting the biometric data through the bound camera. If there is a biometric matching the biometric feature to be identified in the second edge biometric database, the identification is indicated If successful, the second edge node needs to broadcast the biometric feature to the adjacent first edge node, so that the first edge node can receive the biometric feature and decide whether to store it; if it does not exist in the second edge biometric database The biological characteristics that match the biological characteristics to be recognized indicate that the recognition has failed. The second edge node needs to send the biological characteristics to the central node 520 in the cloud environment and request access to the central biological characteristic database 521 deployed in the cloud environment. The feature database 521 stores all the facial features in the edge biometric database 511 in an area. In other words, each edge biometric database is equivalent to a subset of the central biometric database. If the central biometric database 521 is If there is a biological feature that matches the biological feature to be recognized collected by the second edge node, then the matched biological feature is returned to the second edge node, and the second edge node stores it in the second edge biological feature database and then forwards adjacent ones. The first edge node broadcasts the biological characteristic; if there is no biological characteristic matching the biological characteristic to be recognized collected by the second edge node in the central biological characteristic database 521, it means that the biological characteristic appears for the first time and needs to be supplemented To the central biometric database 521.

It should be noted that the establishment of the central biometric database 521 and the addition of newly emerging biometrics to the central biometric database 521 can be manually input into the central node in the cloud environment.

In addition, before the second edge node sends broadcast information to an adjacent edge node, it needs to obtain address information of an edge node whose distance from the second edge node is less than a preset distance. Optionally, the edge node location information management unit 523 in the central node 520 stores the geographic location information and address information of all edge nodes, the second edge node sends a request message to the central node 520, and the central node 520 receives the request After the message, the edge node whose geographic location distance from the second edge node is less than the preset distance is queried, and the corresponding address information is returned to the second edge node. After receiving the address information returned by the central computing device, the second edge node sends broadcast information to the adjacent edge node (for example, the first edge node) corresponding to the address information, where the geographic location information may be the latitude and longitude of each edge node The preset distance can be set as required, and this application is not limited to this, and the address information can include an IP address or other information required for sending broadcast information to edge nodes. Alternatively, each edge node stores the geographic location information and corresponding address information of all other edge nodes in the area to which it belongs, and the second edge node directly queries and determines the edge node whose distance is less than the preset distance and sends broadcast information to it.

S602: The first edge node judges whether the first edge biometrics database includes the biometrics in the broadcast information, if it does, execute step S606; if it does not include it, execute step S603.

Specifically, after receiving the biological characteristics broadcast by the second edge node, the first edge node queries the first edge biological characteristic database. If the biological characteristic is already stored in the first edge biological characteristic database, it means that when the biological characteristic appears When within the management range of the first edge node, the first edge node can recognize the biometrics locally, and determine the identity corresponding to the biometrics. There is no need to send the biometrics to the cloud environment for identification, which can improve The response speed of biometric identification reduces the time delay of biometric identification; if the biometric feature is not stored in the first edge biometric database, the first edge node needs further processing to determine whether it is necessary to store the biometric feature in the first edge biometric Feature database.

S603: The first edge node judges whether the probability value of the biometric feature in the broadcast information appearing in the management range of the first edge node is greater than or equal to a preset threshold, if yes, execute step S604; otherwise, execute step S606.

It is easy to understand that a large part of the biometrics appearing in each location is fixed. If this part of the biometrics is stored in the edge biometric database, the process of biometric identification can be realized locally and the response speed can be improved. Therefore, the first edge node needs to further determine whether the probability value of the biological feature appearing within the management range of the first edge node is greater than or equal to the preset threshold after receiving the biological feature broadcast by the second edge node, and the probability value If it is greater than or equal to the preset threshold, the biometric feature is stored in the first marginal biometric feature database. The preset threshold can be set according to actual needs, for example, it can be set to 0.5, which is not limited in this application.

In a possible implementation, the first edge node queries the cloud manager for the social relationship information corresponding to the first biometric feature (that is, the biometric feature broadcast by the second edge node); the first edge node corresponds to the first biometric feature Calculate the degree of association between the first biometric feature and the first edge node; the first edge node determines whether the value of the degree of association is greater than or equal to a preset threshold.

Specifically, the cloud manager deployed in the cloud environment includes a social relationship library, and the social relationship library may be the social relationship library storage unit 522 shown in FIG. 5. The social relationship database 522 stores the social relationship information corresponding to all the biological characteristics, and the first edge node can calculate the correlation degree indicating the probability that the first biological characteristic appears within the management range of the first edge node according to the social relationship information. , And finally make judgments and follow-up processing based on the calculation results.

In a specific embodiment, the first edge node reads the biometric features in the first edge face database, and the first edge node can read all or part of the biometric features stored in the first edge biometric database, and Query the cloud manager for the social relationship intimacy between the first biological feature and the biological feature in the first edge biological feature database; the first edge node queries the cloud manager for personal tag information and activity track information corresponding to the first biological feature.

It should be noted that, for each biological feature, the interaction between it and other biological features, friendships, etc. can be represented by the intimacy of social relations, and the value of intimacy can be a value between 0-1 , When the value of intimacy is closer to 1, it means that the interaction between the two biological features is more frequent, and the more likely to be friends; when the value of intimacy is closer to 0, it means that there is almost no interaction between the two biological features. The more likely it is a stranger. In addition, there is a corresponding personal tag for each biological feature, which is used to indicate the hobby corresponding to the biological feature. For example, when the personal tag corresponding to a certain biological feature is coffee, it means that the biological feature prefers to drink. Coffee may frequent the coffee shop; when the personal label corresponding to a certain biometric is red wine, it means that the biometric prefers to drink red wine and may frequent the winery. In addition, each biological feature has an activity trajectory, that is, the geographic range of the biological feature's activity (appearance). The general activity trajectory means that the biological feature moves within a range with a certain geographic location as the center and a fixed radius.

It should be understood that the social relationship intimacy, personal tags, and activity trajectories of each biometric feature with other biometrics are also cached in the cloud environment when the central biometric database is constructed, which can be manually entered or through other biometrics. The method of obtaining and storing is not limited in this application.

Further, in addition to the first biological feature, the broadcast information sent by the second edge node also includes geographic location information of the second edge node and label information of the second edge node. The geographic location information of the second edge node may be the latitude and longitude of the geographic location where the second edge node is located; the label information of the second edge node is used to indicate the type of the second edge node, for example, it may be western restaurant, steak, coffee, etc. After receiving the broadcast information sent by the second edge node, the first edge node calculates the first label similarity according to the label information of the first edge node and the label information of the second edge node. The first edge node calculates the distance between the two edge nodes according to the geographic location information of the first edge node and the geographic location information of the second edge node, and then the activity radius corresponding to the first biological feature obtained by the query according to the distance, Calculate geographic similarity. The first edge node calculates the second label similarity according to the label information of the first edge node and the personal label information corresponding to the first biological feature obtained by the query. The first edge node calculates the first biological feature and the first edge according to any one or more of the social relationship intimacy corresponding to the first biological feature, the first tag similarity, the second tag similarity, and the geographic location similarity. The degree of relevance of the node.

It is easy to understand that the higher the affinity between the first biological feature and the biological feature in the first edge biometric database, the greater the probability that the first biological feature appears in the range managed by the first edge node; when the first edge The higher the similarity between the label information of the node and the label information of the second edge node, the greater the probability that the first biological feature appears in the range managed by the first edge node. For example, when the first edge node’s The label is a western restaurant, and the label of the second edge node is steak, then the first biological feature that appears in the management range of the second edge node is very likely to appear in the management range of the first edge node; when the label of the first edge node The higher the similarity between the information and the personal tag information corresponding to the first biometric feature, the greater the probability that the first biometric feature appears within the management range of the first edge node; when the first edge node and the second edge node are between The higher the similarity between the distance of and the radius of activity corresponding to the first biological feature, the greater the probability that the first biological feature appears within the management range of the first edge node.

In a possible implementation manner, the first edge node uses a weighted average algorithm to weight the calculated social relationship intimacy, first label similarity, second label similarity, and geographic location similarity corresponding to the first biological feature. The sum is obtained to obtain the degree of association between the first biological feature and the first edge node.

Specifically, the weight of each dimension can be set according to actual needs, and the sum of all weight factors is 1. As an example, the correlation between the first biological feature and the first edge node can be calculated by the following formula 1:

I=D*W1+S*W2+P*W3+L*W4 Formula 1

Among them, I represents the degree of association between the first biological feature and the first edge node, D represents the similarity of geographic location, S represents the degree of intimacy of social relationships, P represents the similarity of the second label, and L represents the similarity of the first label. W1, W2, W3, and W4 respectively represent weighting factors, and their sum is 1, for example, W1=0.4, W2=0.3, W3=0.2, W1=0.1.

Optionally, the value of D can be calculated using the following formula 2:

D=(1-|A-R|)/(A+R) Formula 2

Among them, A represents the distance between the first edge node and the second edge node, and R represents the radius of activity corresponding to the first biological feature. It should be understood that the value of D can also be calculated in other ways, which is not limited in this application.

The value of S can be calculated in the following way: sort the intimacy between the biometric feature in the first edge biometric database and the first biometric feature according to the value; according to the intimacy value, from high to low Select a fixed number of intimacy values; average the selected fixed number of intimacy values, and finally get S. It should be noted that the fixed number can be set according to needs. If the number of face features in the first edge biometric database is less than the fixed number to be selected, it will be less than part of the intimacy with the first biometric. The value of is recorded as 0. For example, there are 5 biological characteristics in the first edge biological characteristic database, and the fixed number of intimacy values that need to be selected is 6, then when calculating S, the 5 biological characteristics need to be intimacy with the first biological characteristic. Add the values of degrees, and then divide by 6 to get S. It should be understood that the value of S can also be calculated in other ways, which is not limited in this application.

Optionally, L and P can be cosine similarity, that is, the label information of the first edge node is calculated by converting the label information of the first edge node and the second edge node and the personal label information corresponding to the first biometric into a vector The cosine value of the angle between the corresponding vector and the vector corresponding to the label information of the second edge node to obtain L; calculate the angle between the vector corresponding to the label information of the first edge node and the vector corresponding to the personal label The cosine value, so as to get P. The value range of cosine similarity is {-1,1}. The closer to 1, the closer the angle between the two vectors is to 0°, the more similar the two vectors; on the contrary, the closer to -1, then It shows that the closer the angle between the two vectors is to 180°, the two vectors are not similar. It should be understood that the values of L and P can also be calculated in other ways, which are not limited in this application.

It should be noted that the above embodiment comprehensively considers multiple factors, thereby calculating the intimacy between the first biological feature and the first edge node, that is, multi-dimensional calculation of intimacy, of course, it is also possible to consider only one or part of the dimensions ( That is, a combination of several dimensions), for example, only one of geographic location similarity, first tag similarity, second tag similarity, and social relationship intimacy is considered, or any combination of them. The specific implementation process and logic are the same as the above Consistent, for the sake of brevity, I will not repeat them here.

S604: The first edge node stores the biological feature in the first edge biological feature database.

Specifically, after the first edge node calculates the degree of association between the first biological feature and the first edge node, it determines the relationship between the value of the degree of association and the preset threshold. If the value of the degree of association is greater than or equal to the preset threshold, then It shows that the probability that the first biological feature appears in the management range of the first edge node is relatively large, and the first edge node needs to store it in the first edge biological feature database. The preset threshold can be set as required, for example, it can be set to 0.5. When I is greater than or equal to 0.5, the first edge node needs to store the first biological feature.

In a possible implementation manner, when the storage resource of the first edge node is less than a preset storage threshold, the first edge node deletes part of the biological characteristics according to a preset rule.

Specifically, the storage resources of the first edge node are limited and cannot store too many biometrics. When the storage resources are less than the preset storage threshold, some biometrics need to be deleted to ensure that sufficient storage resources are reserved for storing new biometrics. (For example, the first biological feature broadcast by the second edge node).

Optionally, sort the number of occurrences of each biological feature in the most recent period of time (for example, one month) according to the size, and delete the biological features with the number of occurrences less than the first threshold, that is, the biological features with a larger number of occurrences are preferentially saved; Or, sort the similarity values of the personal tag information corresponding to each biometric feature and the tag information of the first edge node according to the magnitude, and delete the biometric feature corresponding to the similarity less than the second threshold; or yes, Sort the last appearance time of each biometric feature in a sequence, and delete the biometric feature whose last appearance time exceeds the third threshold from the current time; or delete part of the biometric feature according to other conditions. In addition, the first edge node can also consider the above multiple factors at the same time to decide which biometrics to delete, for example, consider the number of appearances and the last time of each biometric at the same time; consider the number of appearances of each biometric and the first The similarity of the label information of the edge node; also consider the number of appearances of each biological feature, the last appearance time and the similarity with the label information of the first edge node.

It should be understood that the preset rules can be set as required, and this application does not limit which rules are specifically selected by the first edge node. At the same time, the first threshold, the second threshold, and the third threshold are also set as needed.

It can be understood that when the storage resources are insufficient, the first edge node deletes part of the biometrics according to a preset rule, ensuring that the first biometrics broadcast by the second edge node can be stored, and improving the recognition speed of the first biometrics.

S605: The first edge node uses the first edge biological feature database to identify the biological features collected within the management range of the first edge node.

Specifically, after the first edge node stores the first biometrics broadcast by the second edge node in the first edge biometric database, the first edge node can use the first edge biometric database to manage the first edge node Recognize the biometrics collected within the range. Particularly, when the first biological feature appears in the management range of the first edge node, since the first biological feature is stored in the first edge biological feature database, the first edge node can complete the first biological feature locally. A biometric identification does not need to upload the first biometric to the cloud environment again. The identification is completed by the central node in the cloud environment, which greatly improves the response speed of biometric identification and shortens the time delay of biometric identification. Ensure the real-time and stability of biometric identification.

In a possible implementation manner, the first edge node queries the cloud manager for address information of a third edge node whose distance from the first edge node is less than a preset distance. Optionally, the first edge node sends a request message to the edge node location information management unit 523 to request the address information of the third edge node, and the first edge node sends to the third edge node according to the address information of the third edge node. The second broadcast information, the second broadcast information includes a second biological feature.

It is easy to understand that the first edge node not only needs to receive the biometrics broadcast by neighboring edge nodes and determine whether it needs to be stored in the first edge biometric database, at the same time, the first edge node also needs to broadcast the collected biometrics to other neighboring edge nodes. The second biometric feature so that other adjacent edge nodes can receive the second biometric feature broadcast by the first edge node, and determine whether it needs to be stored in the edge biometric database. The first edge node may obtain the address information of the third edge node from the edge node location information management unit 523, or the first edge node itself may store the geographic location information and address information of the third edge node, and directly query to obtain the third edge node. Address information of the edge node. Other adjacent edge nodes (for example, the third edge node) are similar to the first edge node, and the processing flow after receiving the second biological feature is the same as that of the first edge node. For the sake of brevity, it will not be repeated here.

S606: The first edge node abandons storing the first face feature.

The foregoing describes the methods of the embodiments of the present application in detail. In order to facilitate better implementation of the above solutions of the embodiments of the present application, correspondingly, the following also provides related equipment for cooperating with the implementation of the foregoing solutions.

The embodiment of the present application also provides an edge node, such as the edge node 510 in FIG. 5, which is used to perform the aforementioned biometric identification method. This application does not limit the division of the functional units of the edge node, and each unit in the edge node can be added, reduced, or merged as required. Figure 5 exemplarily provides a division of functional units:

The edge node 510 includes an edge biometric database 511, a biometric extraction unit 512, a biometric recognition unit 513, and a biometric management unit 514.

Specifically, the biological feature extraction unit 512 is configured to perform the foregoing step S601, and optionally perform optional methods in the foregoing steps, to obtain the first biological feature.

The biometrics management unit 514 is used to perform the foregoing steps S602-S604 and step S606, and optionally perform optional methods in the foregoing steps, to calculate the degree of association between the first biometrics and the edge node, and to determine whether to store it In the marginal biometric database 511.

The biometric identification unit 513 is configured to perform the foregoing step S605, and optionally perform optional methods in the foregoing steps, using the edge biometric database 511 to recognize the biometrics collected within the management range of the edge node.

The above four units can transmit data to each other through a communication path. It should be understood that each unit included in the edge node 510 can be a software unit, a hardware unit, or a part of a software unit and a part of a hardware unit.

Refer to FIG. 7, which is a schematic structural diagram of a computing device provided by an embodiment of the present application. As shown in FIG. 7, the computing device 700 includes a processor 710, a communication interface 720, and a memory 730. The processor 710, the communication interface 720, and the memory 730 are connected to each other through an internal bus 740. It should be understood that the computing device may be a general-purpose server.

The processor 710 may be composed of one or more general-purpose processors, such as a central processing unit (CPU), or a combination of a CPU and a hardware chip. The aforementioned hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above-mentioned PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a general array logic (generic array logic, GAL), or any combination thereof.

The bus 740 may be a peripheral component interconnect standard (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 740 can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is used in FIG. 7, but it does not mean that there is only one bus or one type of bus.

The memory 730 may include a volatile memory (volatile memory), such as a random access memory (random access memory, RAM); the memory 730 may also include a non-volatile memory (non-volatile memory), such as a read-only memory (read-only memory). Only memory, ROM, flash memory, hard disk drive (HDD), or solid-state drive (SSD); the memory 730 may also include a combination of the above types. The program code may be used to implement the functional modules shown in the edge node 510, or to implement the method steps in the method embodiment shown in FIG. 6 with the first edge node as the execution subject.

The embodiments of the present application also provide a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it can implement part or all of the steps of any one of the above method embodiments, and realize the above The function of any one of the functional modules described in Figure 5.

As shown in FIG. 8, the present application also provides a computing device cluster, and the computing device cluster includes a plurality of computing devices 800. The organizational structure of each computing device 800 is the same as that of the computing device 700, and includes a processor 810, a communication interface 820, and a memory 830. The processor 810, the communication interface 820, and the memory 830 are connected to each other through an internal bus 840.

Each computing device 800 establishes a communication path through a communication network. Each computing device 800 runs any one or more of the edge biometric database 511, the biometric extraction unit 512, the biometric identification unit 513, and the biometric management unit 514. Any computing device 800 may be a computing device in an edge computing device system, or a terminal computing device.

The embodiments of the present application also provide a computer program product, which when it runs on a computer or a processor, enables the computer or the processor to execute one or more steps in any of the foregoing methods. If each component module of the aforementioned equipment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the computer readable storage medium.

In the above-mentioned embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in an embodiment, reference may be made to related descriptions of other embodiments.

It should be understood that the first, second, third, fourth, and various numerical numbers involved in this specification are only for easy distinction for description, and are not used to limit the scope of the present application.

It should be understood that the term "and/or" in this text is only an association relationship describing the associated objects, indicating that there can be three types of relationships, for example, A and/or B, which can mean: A alone exists, and both A and B exist. , There are three cases of B alone. In addition, the character "/" in this text generally indicates that the associated objects before and after are in an "or" relationship.

It should also be understood that, in the various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not be implemented in this application. The implementation process of the example constitutes any limitation.

A person of ordinary skill in the art may be aware that the units and algorithm steps of the examples described in the embodiments disclosed in this document can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

Those skilled in the art can clearly understand that, for the convenience and conciseness of the description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiment, which is not repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device, and method may be implemented in other ways. For example, the device embodiments described above are merely illustrative, for example, the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or It can be integrated into another system, or some features can be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.

If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solution of the present application essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

The steps in the method in the embodiment of the present application can be adjusted, merged, and deleted in order according to actual needs.

The modules in the devices in the embodiments of the present application may be combined, divided, and deleted according to actual needs.

As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions recorded in the embodiments are modified, or some of the technical features are equivalently replaced; these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

A method of biometric identification, which is characterized in that it includes:

The first edge node receives the broadcast information sent by the second edge node, the broadcast information includes a first biometric feature, the first edge node stores a first edge biometric feature database, and the first edge biometric feature database includes at least one biometric feature. feature;

It is determined that the probability value of the first biological feature appearing in the range managed by the first edge node is greater than or equal to a preset threshold, and the first edge node stores the first biological feature in the first edge biological feature Library

The first edge node uses the first edge biometrics database to identify the biometrics collected within the management range of the first edge node.
The method of claim 1, wherein the first edge node determining that the probability value of the first biometric feature appearing in the range managed by the first edge node is greater than or equal to a preset threshold comprises:

The first edge node queries a cloud manager for social relationship information corresponding to the first biological characteristic, and the cloud manager includes a social relationship database, and the social relationship database includes social relationship information corresponding to the biological characteristics;

The first edge node calculates the degree of association between the first biometric feature and the first edge node according to the social relationship information corresponding to the first biometric feature, and the degree of association is used to indicate the first biometric feature Probability of appearing within the management range of the first edge node;

The first edge node determines that the value of the degree of association is greater than or equal to the preset threshold.
3. The method of claim 2, wherein the first edge node querying a cloud manager for social relationship information corresponding to the first biometric feature comprises:

The first edge node reads the biological characteristics in the first edge biometric database, and queries the cloud manager for the social characteristics of the first biological characteristics and the biological characteristics in the first edge biometric database. Relationship intimacy

The first edge node queries the cloud manager for personal tag information and activity track information corresponding to the first biometric feature.
The method according to claim 3, wherein the broadcast information further includes geographic location information of the second edge node and label information of the second edge node; the first edge node calculates the first edge node The degree of association between a biological feature and the first edge node includes:

The first edge node calculates a first label similarity according to the label information of the first edge node and the label information of the second edge node, and the label information of the first edge node is used to indicate the first edge node. The type of the edge node, where the label information of the second edge node is used to indicate the type of the second edge node;

The first edge node calculates the distance between the first edge node and the second edge node according to the geographic location information of the first edge node and the geographic location information of the second edge node, and according to the distance and the location information. The activity track information calculates the similarity of geographic location;

The first edge node calculates a second tag similarity according to the tag information of the first edge node and the personal tag information corresponding to the first biometric feature;

The first edge node calculates the first creature according to any one or more of the social relationship intimacy, the first tag similarity, the second tag similarity, and the geographic location similarity. The degree of association between the feature and the first edge node.
The method according to any one of claims 1-4, wherein the method further comprises:

When the storage resource of the first edge node is less than a preset storage threshold, the first edge node deletes the biometric feature according to a preset rule;

The preset rules include:

Delete the biometrics whose appearances are less than the first threshold; or,

Delete the biological features whose similarity with the label information of the first edge node is less than the second threshold; or,

Delete the biological feature whose last appearance time exceeds the third threshold from the current time.
The method according to any one of claims 1-5, wherein the method further comprises:

The first edge node queries a cloud manager for address information of a third edge node whose distance from the first edge node is less than a preset distance, and the cloud manager includes an edge node information database, and the edge node information database Including the geographic location information and address information of the edge node;

The first edge node sends second broadcast information to the third edge node according to the address information of the third edge node, where the second broadcast information includes a second biological characteristic.
An edge node, characterized in that it includes:

Edge biometric database, used to store at least one biometric feature;

The biological feature extraction unit is configured to receive the broadcast information of the second edge node and extract the broadcast information including the first biological feature;

A biometrics management unit, configured to determine that the probability value of the first biometrics appearing in the range managed by the edge node is greater than or equal to a preset threshold, and store the first biometrics in the edge biometrics database;

The biological feature recognition unit is configured to use the edge biological feature database to recognize the biological features collected within the management range of the edge node.
The edge node according to claim 7, wherein:

The biological feature management unit is also used for:

Query a cloud manager for social relationship information corresponding to the first biological characteristic, where the cloud manager includes a social relationship database, and the social relationship database includes social relationship information corresponding to the biological characteristics;

Calculate the degree of association between the first biometric feature and the edge node according to the social relationship information corresponding to the first biometric feature, where the degree of association is used to indicate that the first biometric feature appears in the management of the edge node Probability within range;

It is determined that the value of the degree of association is greater than or equal to the preset threshold.
The edge node according to claim 8, wherein:

The biological feature management unit is also used for:

Read the biometrics in the edge biometrics database, and query the cloud manager for the social relationship intimacy between the first biometrics and the biometrics in the edge biometrics database;

Query the cloud manager for personal tag information and activity track information corresponding to the first biological feature.
The edge node according to claim 9, wherein the broadcast information further includes geographic location information of the second edge node and label information of the second edge node;

The biological feature management unit is also used for:

The first label similarity is calculated according to the label information of the edge node and the label information of the second edge node. The label information of the edge node is used to indicate the type of the edge node, and the label information of the second edge node Label information is used to indicate the type of the second edge node;

Calculate the distance between the edge node and the second edge node according to the geographic location information of the edge node and the geographic location information of the second edge node, and calculate the geographic location similarity based on the distance and the activity track information ；

Calculating the second tag similarity according to the tag information of the edge node and the personal tag information corresponding to the first biometric feature;

Calculate the first biological feature and the edge node according to any one or more of the social relationship intimacy, the first tag similarity, the second tag similarity, and the geographic location similarity The degree of relevance.
The edge node according to any one of claims 7-10, wherein:

The biological feature management unit is also used for:

When the storage resource of the edge node is less than the preset storage threshold, delete the biometric feature according to the preset rule;

The preset rules include:

Delete the biometrics whose appearances are less than the first threshold; or,

Delete the biometrics whose similarity with the label information of the edge node is less than the second threshold; or,

Delete the biological feature whose last appearance time exceeds the third threshold from the current time.
The edge node according to any one of claims 7-11, wherein:

The biometric management unit is further configured to query the cloud manager for the address information of a third edge node whose distance from the edge node is less than a preset distance, and the cloud manager includes an edge node information database, and the edge node The information database includes geographic location information and address information of edge nodes;

The biological feature management unit is further configured to send second broadcast information to the third edge node according to the address information of the third edge node, where the second broadcast information includes the second biological feature.
A computing device cluster, wherein the computing device cluster includes at least one computing device, each computing device includes a memory and a processor, and the processor of the at least one computing device executes the memory storage of the at least one computing device. The computer instructions for causing the at least one computing device to execute the method according to any one of claims 1-6.
A computer storage medium storing a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is realized.
A computer program, the computer program includes instructions, when the computer program is executed by a computer, the computer can execute the method according to any one of claims 1-6.