CN111221991B

CN111221991B - Method and device for determining personnel identity attribute and electronic equipment

Info

Publication number: CN111221991B
Application number: CN201911079417.0A
Authority: CN
Inventors: 程皓; 刘瑞伟
Original assignee: Beijing Kuangshi Technology Co Ltd
Current assignee: Beijing Kuangshi Technology Co Ltd
Priority date: 2019-11-06
Filing date: 2019-11-06
Publication date: 2024-06-14
Anticipated expiration: 2039-11-06
Also published as: CN113505251A; CN111221991A

Abstract

The invention provides a method and a device for determining personnel identity attributes and electronic equipment, wherein the method comprises the following steps: acquiring a portrait file of a person to be identified; determining archive labels and/or basic attribute information according to the portrait archive; and determining the personnel identity attribute of the personnel to be identified corresponding to the personnel image file according to the corresponding relation between the file label and/or the basic attribute information and the personnel identity attribute. The method can carry out multidimensional analysis on the portrait file to obtain file labels and/or file basic attribute information, and determines the personnel identity attribute of the personnel to be identified corresponding to the portrait file based on the corresponding relation between the file labels and/or the basic attribute information and the personnel identity attribute, thereby relieving the technical problem that the identity attribute of the personnel is difficult to effectively determine in the prior art.

Description

Method and device for determining personnel identity attribute and electronic equipment

Technical Field

The present invention relates to the technical field of personnel identity recognition, and in particular, to a method and an apparatus for determining a personnel identity attribute, and an electronic device.

Background

The human face recognition system covering the public area and the vital parts is gradually constructed in China, and people passing by can be subjected to face snapshot. Aiming at the established face recognition system, real-time face snap shots can be aggregated through a face clustering technology, so that a face file (a collection of snap shots of the same person in essence) of each person is generated. When determining the identity attribute of a person corresponding to a portrait file, the identity attribute of the person corresponding to the portrait file can be determined by matching the person real-name photo library with the portrait file, which is often needed to rely on the person real-name photo library and the person identity attribute provided by related departments.

In a practical application environment, it is difficult for the relevant departments to provide all or a large number of real-name photo libraries, and even if the relevant departments provide the real-name photo libraries of all the mastered people, the relevant departments cannot completely match a large number of portrait files, because a large number of unconscious people (such as floating population and people not registering real information) exist, and the unconscious people are just people needing important attention. In addition, in general, only identity information (e.g., name information, age information, occupation information, etc.) of a person is obtained by a manner in which the person names a photo library in real time, but business attribute information of the person cannot be obtained.

To sum up, it is difficult in the prior art to effectively determine identity attributes of personnel.

Disclosure of Invention

In view of the above, the present invention aims to provide a method, an apparatus and an electronic device for determining identity attributes of a person, so as to alleviate the technical problem that it is difficult to effectively determine identity attributes of a person in the prior art.

In a first aspect, an embodiment of the present invention provides a method for determining a person identity attribute, which is applied to a person analysis system, including: acquiring a portrait file of a person to be identified; the portrait file at least comprises: the system comprises a portrait snapshot, snapshot time of the portrait snapshot and information of an image acquisition device for snapshot of the portrait snapshot, wherein each portrait file corresponds to a person to be identified; determining archive labels and/or basic attribute information according to the portrait archive; and determining the personnel identity attribute of the personnel to be identified corresponding to the personnel image file according to the corresponding relation between the file label and/or the basic attribute information and the personnel identity attribute.

Further, the archive tag includes: at least one of an abnormal behavior type file tag, a personnel state type file tag, a situation type file tag and an affinity type file tag; the archive tag of the abnormal behavior category comprises: daytime and night out labels, frequent out labels, personnel gathering labels and regional resident labels; the profile tag of the personnel status category includes: a resident tag, an leave district tag, and a visit area tag; the archive label of the present situation category comprises: in the preset time, labels which do not reach the preset times are displayed in the same image acquisition device or the same area; the archive label of the affinity class comprises: a label of the same-line times of the person with the known person identity attribute; the step of determining the archive label and/or the basic attribute information according to the portrait archive comprises the following steps: and carrying out statistical analysis on the portrait file to obtain the file label and/or the basic attribute information.

Further, the step of determining the personnel identity attribute of the personnel to be identified corresponding to the portrait file according to the corresponding relationship between the archive label and/or the basic attribute information and the personnel identity attribute includes: determining a target portrait file in a plurality of portrait files according to preset basic attribute information; and matching the archive label of the target portrait archive with the archive label corresponding to the preset basic attribute information in the corresponding relation, and determining the personnel identity attribute of the personnel to be identified corresponding to the target portrait archive according to the matching result.

Further, after obtaining the personnel identity attribute of the personnel to be identified corresponding to the portrait file, the method further includes: dividing a plurality of portrait files with the personnel identity attribute determined to obtain a portrait file set with personnel identity attribute information; taking the personnel identity attribute information of the personnel image file set as the group identity attribute of the personnel group corresponding to the personnel image file set; and/or carrying out multidimensional analysis on the portrait files in the portrait file set to obtain group characteristics of the personnel group; the multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, location dimension analysis, and affinity dimension analysis.

Further, the method further comprises: performing group identification on personnel in the target personnel group corresponding to the target portrait archive set based on group discovery conditions to obtain personnel groups; the target set of portrait files is a subset of the set of portrait files.

Further, the subset is a portrait file set matched with a preset group identity attribute and/or a preset group characteristic in the portrait file set.

Further, the step of performing group identification on the people in the target people group corresponding to the target portrait archive set based on the group discovery condition includes: determining the intimacy between target persons in the target person group; the target personnel are any two personnel in the target personnel group; if the affinity between the target persons is greater than a preset affinity threshold, determining that the target persons are relational persons; building a relationship chain by taking the relational personnel as nodes; determining a target relationship chain among the relationship chains; the target relationship chain is any one of the relationship chains; and if the number of the nodes of the target relation chain is larger than a preset threshold, taking the relational personnel corresponding to the nodes of the target relation chain as personnel in the personnel group, and further obtaining the personnel group.

Further, the method further comprises: determining the intimacy between the personnel in the personnel group and the personnel outside the personnel group based on the personnel profile of the personnel in the personnel group and the personnel outside the personnel group; and determining the edge personnel of the personnel group according to the intimacy.

Further, the method further comprises: carrying out multidimensional analysis on the portrait files corresponding to the personnel groups to obtain group characteristics of the personnel groups; the multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, place dimension analysis and affinity dimension analysis; and determining a target personnel group in the personnel groups according to the preset group characteristics.

In a second aspect, an embodiment of the present invention further provides a device for determining a person identity attribute, which is applied to a person analysis system, including: the acquisition unit is used for acquiring the portrait file of the person to be identified; the portrait file at least comprises: the system comprises a portrait snapshot, snapshot time of the portrait snapshot and information of an image acquisition device for snapshot of the portrait snapshot, wherein each portrait file corresponds to a person to be identified; the first determining unit is used for determining archive labels and/or basic attribute information according to the portrait archive; and the second determining unit is used for determining the personnel identity attribute of the personnel to be identified corresponding to the portrait file according to the corresponding relation between the archive label and/or the basic attribute information and the personnel identity attribute.

In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the steps of the method according to any one of the first aspects when the processor executes the computer program.

In a fourth aspect, embodiments of the present invention provide a computer storage medium having stored thereon a computer program which, when run, performs the steps of the method of any of the first aspects described above.

In the embodiment of the invention, a portrait file of a person to be identified is firstly obtained; then, determining archive labels and/or basic attribute information according to the portrait archive; and finally, determining the personnel identity attribute of the personnel to be identified corresponding to the personnel image file according to the corresponding relation between the file label and/or the basic attribute information and the personnel identity attribute. As can be seen from the above description, the method according to the embodiment of the present invention can perform multidimensional analysis on the portrait file to obtain the file tag and/or the file base attribute information, and determine the personnel identity attribute of the personnel to be identified corresponding to the portrait file based on the corresponding relationship between the file tag and/or the base attribute information and the personnel identity attribute, thereby alleviating the technical problem that it is difficult to effectively determine the personnel identity attribute in the prior art.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of an electronic device according to an embodiment of the present invention;

FIG. 2 is a flowchart of a method for determining a person identity attribute according to an embodiment of the present invention;

FIG. 3 is a flowchart of a method for determining a person identity attribute of a person to be identified corresponding to a portrait file according to an embodiment of the present invention;

FIG. 4 is a flowchart of a method for identifying groups of people in a target group of people corresponding to a target set of portrait files according to an embodiment of the present invention;

FIG. 5 is a flowchart of a method for community screening according to an embodiment of the present invention;

Fig. 6 is a schematic diagram of a device for determining a person identity attribute according to an embodiment of the present invention.

Detailed Description

The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Example 1:

First, an electronic device 100 for implementing an embodiment of the present invention, which may be used to run the method for determining a person identity attribute of embodiments of the present invention, will be described with reference to fig. 1.

As shown in fig. 1, electronic device 100 includes one or more processors 102, one or more memories 104, an input device 106, an output device 108, and a camera 110, which are interconnected by a bus system 112 and/or other forms of connection mechanisms (not shown). It should be noted that the components and structures of the electronic device 100 shown in fig. 1 are exemplary only and not limiting, as the electronic device may have other components and structures as desired.

The processor 102 may be implemented in hardware in at least one of a digital signal processor (DSP, digital Signal Processing), field-Programmable gate array (FPGA), programmable logic array (PLA, programmable Logic Array), and ASIC (Application SPECIFIC INTEGRATED Circuit), the processor 102 may be a central Processing unit (CPU, central Processing Unit) or other form of Processing unit having data Processing and/or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.

The memory 104 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that can be executed by the processor 102 to implement client functions and/or other desired functions in embodiments of the present invention as described below. Various applications and various data, such as various data used and/or generated by the applications, may also be stored in the computer readable storage medium.

The input device 106 may be a device used by a user to input instructions and may include one or more of a keyboard, mouse, microphone, touch screen, and the like.

The output device 108 may output various information (e.g., images or sounds) to the outside (e.g., a user), and may include one or more of a display, a speaker, and the like.

The camera 110 is configured to collect a person image file of a person to be identified, where the person image file of the person to be identified collected by the camera is processed by the method for determining a person identity attribute to obtain a person identity attribute of the person to be identified, for example, the camera may capture a person image expected by a user, and then process the person image by the method for determining a person identity attribute to obtain a person identity attribute of a person image corresponding to the person, and the camera may store the captured person image in the memory 104 for use by other components.

By way of example, the electronic device for implementing the method for determining the identity attribute of a person according to the embodiment of the present invention may be implemented as a smart mobile terminal such as a smart phone, a tablet computer, or any other device having computing capabilities.

Example 2:

according to an embodiment of the present invention, there is provided an embodiment of a method of determining a person identity attribute, it being noted that the steps shown in the flowchart of the figures may be performed in a computer system such as a set of computer executable instructions, and, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order different from that shown or described herein.

Fig. 2 is a flowchart of a method for determining a person identity attribute according to an embodiment of the present invention, as shown in fig. 2, the method includes the steps of:

step S202, a portrait file of a person to be identified is obtained.

In the embodiment of the invention, the method for determining the identity attribute of the person can be applied to a person analysis system, the person analysis system can be in butt joint with a person image big data system, and the person image big data system is used for clustering the person image snap shots of the person to be identified, which are acquired by the image acquisition device, into the person image files according to the corresponding person, and further, the person analysis system acquires the person image files of the person to be identified. The person to be identified may be a person whose identity information is known but whose service attribute information is unknown, or may be a person whose identity information and service attribute information are both unknown. The identity information refers to name information, age information, identification card number information, occupation information and the like; and the service attribute information refers to information related to a specific service. For example, the service attribute information may be r service attribute, s service attribute, t service attribute, or the like.

The portrait file at least comprises: the portrait snapshot, the snapshot time of the portrait snapshot and the information of the image acquisition device for snapshot of the portrait snapshot, besides the above information, the portrait file can also comprise: the personal photo, the name corresponding to the personal photo, the identification card number corresponding to the personal photo, the age corresponding to the personal photo, the motor vehicle frequently driven by the person to be identified, the mobile phone number of the person to be identified, the information (such as IMEI code, MAC address and RFID signal) acquired by the probe, and the like.

It should be noted that the number of the portrait files may be plural or one, and when the number of the portrait files is plural, each portrait file corresponds to one person to be identified.

Step S204, the archive label and/or the basic attribute information are determined according to the portrait archive.

After the portrait file is obtained, statistical analysis and/or basic attribute analysis are performed on the portrait file. The statistical analysis may include time dimension analysis of the snapshot time of the portrait snapshot, location dimension analysis of the snapshot location of the portrait snapshot (which may be determined by the information of the image collecting device), and affinity dimension analysis of the affinities of the people corresponding to different portrait files, and through the statistical analysis, features (such as behavior rule features, activity rule features, and features of the interdynamic people) of the people corresponding to the portrait files may be obtained, and these features may be used as file labels of the portrait files corresponding to the people to be identified. In addition, the basic attribute analysis includes analyzing basic attributes of all the portrait candid graphs in the portrait file to obtain basic attribute information (such as gender information, age information, race information, whether to drive or not) of the person to be identified corresponding to the portrait file. The analysis of the underlying properties of the archive may be performed by a neural network model.

It should be noted that a portrait file may correspond to a plurality of file tags, and each file tag may represent a feature of a person to be identified corresponding to the portrait file. In addition, the process of obtaining the archive label through the statistical analysis can be updated according to the setting of the user. For example, before updating, if the number of the snap shots of the person a shot at night is greater than the number of the snap shots of the person a shot in the daytime, determining that one of the archive labels of the person a is a label of daytime and night; after updating, the number of the snap shots of the A person shot at night is required to be statistically analyzed to be larger than the number of the snap shots of the A person shot in the daytime, and when the number of the snap shots of the A person shot in the daytime is smaller than a preset threshold value, one tag in the archive tag of the A person is determined to be a tag which is in daytime and nighttime. The process of determining the archive label can be added or changed according to the service requirement, and the process of determining the archive label is not limited by the invention.

Step S206, according to the corresponding relation between the archive label and/or the basic attribute information and the personnel identity attribute, the personnel identity attribute of the personnel to be identified corresponding to the portrait archive is determined.

When the method is implemented, the personnel identity attribute of the personnel to be identified corresponding to the portrait file can be determined only according to the corresponding relation between the file label and the personnel identity attribute; the personnel identity attribute of the personnel to be identified corresponding to the personnel image file can be determined only according to the corresponding relation between the basic attribute information and the personnel identity attribute, and the personnel identity attribute of the personnel to be identified corresponding to the personnel image file can be determined according to the corresponding relation between the file label and the basic attribute information and the personnel identity attribute, and the personnel identity attribute of the personnel to be identified corresponding to the personnel image file can be specifically selected according to the requirement.

In addition, the corresponding relations are multiple, and each corresponding relation is the correspondence between the file label and/or the permutation and combination of the basic attribute information and one personnel identity attribute.

The permutation and combination may be an intersection of a plurality of archive labels and/or underlying attribute information. For example, the basic attribute information corresponding to the r service attribute (identity attribute) is: male, normally open car, and the archive label that r business attribute corresponds is: frequent departure from a train station and daytime and night departure, namely when the archive label of the portrait archive simultaneously comprises frequent departure from the train station and daytime and night departure, and basic attribute information corresponding to the portrait archive is a male service attribute and a normally open service attribute, determining that the personnel identity attribute of the personnel to be identified corresponding to the portrait archive is an r service attribute;

The permutation and combination may be a union of multiple archive labels and/or base attribute information. For example, the archive label corresponding to the person with w service attribute is: the personal identity attribute of the person to be identified corresponding to the portrait file is determined to be w business attribute when the file label of the portrait file comprises a label with high affinity to the person with the s business attribute or a label with high affinity to the person with the v business attribute;

The permutation and combination may also be a complement of the plurality of archive labels and/or the underlying attribute information. For example, the archive labels corresponding to suspicious people are: diurnal emission and within the jurisdiction, that is to say when the archive tag of the portrait archive includes a diurnal emission tag, but does not include a tag leaving the jurisdiction (this means within the jurisdiction), the person identity attribute of the person to be identified corresponding to the portrait archive is determined to be a suspicious person.

It should be noted that, the above-mentioned correspondence may be set manually through experience (specifically, the related personnel can master the activity rules and behavior characteristics of personnel with different identity attributes), or may also be obtained by performing statistical analysis on the portrait files of personnel with known identity attributes to obtain basic attribute information and/or file labels, and performing multidimensional analysis on the groups of personnel with the same identity attributes to determine, where a specific multidimensional analysis means will be described in detail later. The embodiment of the invention does not limit the determination process of the corresponding relation.

The above description briefly describes the method for determining the identity attribute of a person according to the present invention, and the following description refers to specific details.

In this embodiment, determining the base attribute information from the portrait archive includes:

Determining file characteristics of the portrait file according to the portrait snapshot in the portrait file; and determining the basic attribute information of the personnel to be identified corresponding to the portrait file according to the file characteristics.

The file features are used for representing the portrait features of the personnel to be identified corresponding to the portrait snapshot; the basic attribute information includes: at least one of sex information, age information, race information, and whether to drive.

Considering that the portrait snapshot in the portrait file can reflect the portrait characteristics of the person to be identified, and each portrait snapshot can only reflect part of the portrait characteristics of the person to be identified, the portrait characteristics of all portrait snapshots in the portrait file need to be fused, so that the file characteristics of the portrait file are obtained.

When the method is implemented, the image characteristic of the best-quality portrait snapshot in the portrait snapshot can be used as the file characteristic of the portrait file; the image characteristics of the latest shot portrait snapshot in the portrait snapshot can be used as the archive characteristics of the portrait archive; the comprehensive characteristics of the multiple portrait candid pictures can also be used as portrait archive characteristics. When the comprehensive characteristics of the multiple portrait snapshots are taken as the portrait archival characteristics, for example, a the image quality of the eye part of the portrait snapshot is best, a larger weight can be set for the eye part of the a portrait snapshot, and smaller weights can be set for the eye part of the other portrait snapshots, b the image quality of the nose part of the portrait snapshot is best, a larger weight can be set for the nose part of the b portrait snapshot, and smaller weights can be set for the nose part of the other portrait snapshots, so that weight setting is performed on each part of the portrait snapshots. And further, after obtaining the characteristic values of the plurality of portrait snapshots, carrying out weighted average on the characteristic values of the plurality of portrait snapshots according to the set weight, and finally obtaining a weighted average result which is the comprehensive characteristic. The process of determining the profile characteristics of the portrait file is not particularly limited, and the profile characteristics of the portrait file can be determined in other manners.

In another embodiment, determining the base attribute information from the portrait archive includes: and determining basic attribute information of the person to be identified according to the portrait snapshot in the portrait file.

Specifically, the method comprises the following steps: carrying out attribute identification on each portrait snapshot in the portrait file through an attribute identification model so as to obtain a plurality of basic attribute information corresponding to each portrait snapshot one by one; among the plurality of basic attribute information, basic attribute information of the person to be identified is determined based on the plurality of basic attribute information of the same category.

For example, there are 5 portrait snapshots in the portrait file, after attribute identification is performed on the 5 portrait snapshots through an attribute identification model, 4 portrait snapshots displayed in the 5 portrait snapshots are female, and 1 portrait displayed in the 5 portrait snapshots is male, and meanwhile, the age corresponding to each portrait snapshot is obtained, so that the sex attribute of the person to be identified corresponding to the portrait file can be determined to be female according to the result, and further the age corresponding to the 5 portrait snapshots is averaged to be used as the age attribute of the person to be identified corresponding to the portrait file.

In this embodiment, the archive tag includes: at least one of an abnormal behavior type file tag, a personnel state type file tag, a situation type file tag and an affinity type file tag;

The archive labels of the abnormal behavior categories include: daytime and night out labels, frequent out labels, personnel gathering labels and regional resident labels;

the profile tag for the personnel status category includes: a resident tag, an leave district tag, and a visit area tag;

the archive labels of the show and miss categories include: in the preset time, labels which do not reach the preset times are displayed in the same image acquisition device or the same area;

The archive label of the affinity class includes: a label of the same-line times of the person with the known person identity attribute;

The step of determining the archive label and/or the basic attribute information according to the portrait archive comprises the following steps:

and carrying out statistical analysis on the portrait file to obtain file labels and/or basic attribute information.

The archive tag may strictly include only the information, or may include detailed information such as the time and place. For example, the frequent-occurrence tag may be frequent occurrence, or frequent occurrence at the location X (e.g., in a middle-guan village); the personnel gathering label can be used for gathering personnel only, or gathering N persons at the place X; the label of the same-line number of the personnel with the known personnel identity attribute can be the same-line number of the personnel with the k service attribute for 5 times, or the same-line number of the personnel with the r service attribute for 4 times, and the like, and the embodiment of the invention does not specifically limit the file label.

The daytime and night time out label refers to that the frequency of occurrence at night is larger than the frequency of occurrence at daytime. In the implementation, if a person to be identified a 22:00-6:00 is beaten more than 6:00-22:00 times of shooting, determining that an archive label of the person A to be identified is diurnal;

The frequent-going-out label refers to that the number of going-out times exceeds Z times in a certain area for a certain time. In implementation, if the number of times that a person a to be identified is shot by a camera in a Zhongguan area (the area includes A, B, C, D, E cameras) is greater than Z times in a day, determining that an archive tag of the person a to be identified is frequently found or found in Zhongguan;

the people gathering label refers to that more than M people appear in a certain area for a certain time. In the implementation, if the number of people in the Y area photographed by the camera in the Y area exceeds M people all the time within 3 hours, determining that one archive label of the M people is person aggregation or person aggregation in the Y area;

The above-mentioned area resident tag means that a person is continuously present for a certain time in a certain area. In the implementation process, if the person A to be identified is continuously shot by the camera in the P area for 5 hours, determining that one file label of the person A to be identified resides in the area or resides in the P area;

The resident label refers to that the number of times of occurrence in the set area is more than Q times in a period of time. In the implementation process, if the number of times that the person A to be identified is shot by a camera in the sea lake area in Beijing city is more than Q times within half a year, determining that one file label of the person A to be identified is a normally-resident person or a normally-resident sea lake area person;

The leaving district label refers to that no continuous O days appear in the set district within a period of time. When the method is implemented, if the person A to be identified is not shot by the camera in the L district for one month in continuous O days, determining that one file label of the person A to be identified is leaving the district or leaves the L district;

The above-mentioned frequent regional label means that the number of times of occurrence and expiration exceeds the number of days of R times by more than S days in a certain region for a certain period of time. In implementation, if the number of times that the person a to be identified is photographed by the camera of the lake area exceeds R times is greater than S days in two months, determining that one archive label of the person a to be identified is a constant visit area or a constant visit lake area.

In this embodiment, referring to fig. 3, an example of determining a person identity attribute of a person to be identified corresponding to a portrait file according to a correspondence between a combination of a profile tag and basic attribute information and the person identity attribute is shown, and specifically includes the following steps:

step S301, determining a target portrait file from a plurality of portrait files according to preset basic attribute information.

The preset basic attribute information may be basic attribute information of interest to the user, or basic attribute information related to identity attribute of interest to the user, which may be specifically set according to needs. The above step S301 is mainly used for narrowing the analysis range of the portrait file.

For example, if the basic attribute information is a man and a normally open car, the basic attribute information is determined to be a man and the normally open car is determined to be a target portrait file in the plurality of portrait files.

Step S302, matching the file label of the target portrait file with the file label corresponding to the preset basic attribute information in the corresponding relation, and determining the personnel identity attribute of the personnel to be identified corresponding to the target portrait file according to the matching result.

The target portrait file is determined on the one hand to reduce the analysis range of the portrait file, and on the other hand, the finally determined personnel identity attribute is more accurate.

The process is described below by way of an example:

If the basic attribute information corresponding to the r service attribute (identity attribute) is: male, normally open car, and the archive label that r business attribute corresponds is: if the person frequently goes out of the train station and goes out of the night in daytime, all the portrait files are analyzed, and all the personnel identity attributes of the personnel to be identified corresponding to the portrait files with the daytime-going out of the tag of the train station and the portrait files frequently goes out of the tag of the train station are determined to be the identities of r service attributes, and in fact, the personnel with the daytime-going out of the tag and the personnel frequently goes out of the tag of the train station have not only the r service attributes, but also the u service attributes (such as cleaning personnel). If the target portrait files of men and normally open cars are determined in all portrait files, then the target portrait files are analyzed, and personnel identity attributes of personnel to be identified corresponding to portrait files with daytime and night out labels and frequent in train station labels in file labels are determined to be r service attributes, so that the determined personnel identity attributes are relatively more accurate.

In addition, in the specific implementation, the multiple portrait files can be filtered through preset basic attribute information to obtain a target portrait file, and then the personnel identity attribute of the personnel to be identified is determined according to the file label of the target portrait file; the personal identity attribute can be filtered according to the archive label of the portrait archive to obtain a plurality of personal identity attributes corresponding to the archive label of the portrait archive, and then the unique personal identity attribute is determined according to the preset basic attribute information. For example, possible person identity attributes determined from the archive tag include: and the identity attribute of the person can be determined to be the r service attribute after the basic attribute information of the male and the normally open vehicles is added to the u service attribute (such as cleaning personnel) and the r service attribute.

In this embodiment, after obtaining the personnel identity attribute of the personnel to be identified corresponding to the portrait file, the method further includes the following steps (1) and (2):

(1) Dividing a plurality of portrait files with the personnel identity attribute determined to obtain a portrait file set with personnel identity attribute information;

(2) Taking the personnel identity attribute information of the personnel image file set as the group identity attribute of the personnel group corresponding to the personnel image file set; and/or carrying out multidimensional analysis on the portrait files in the portrait file set to obtain group characteristics of the personnel group.

Wherein the multidimensional analysis comprises at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, location dimension analysis, and affinity dimension analysis.

The person image file with the determined person identity attribute can be the person image file with the person identity attribute determined by the method of the invention at the current moment, can be the person image file with the person identity attribute determined before, can also be the person image file with the known person identity attribute, and the person image file with the determined person identity attribute is not limited in the embodiment of the invention.

When the portrait files are divided, the portrait files corresponding to the same personnel identity attribute are divided together to obtain a portrait file set with personnel identity attribute information, the personnel identity attribute information of the portrait file set is used as group identity attribute of a group corresponding to the portrait file set, and multidimensional analysis is performed on the group corresponding to the portrait file set to obtain group characteristics of the group.

The multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, location dimension analysis, and affinity dimension analysis. Corresponding to the characteristic, the group characteristics comprise basic attribute information, archive labels, appearance time characteristics, location characteristics and peer personnel identity attribute characteristics of personnel in the personnel group, (such as personnel in the personnel group are frequently shot in an area A from 6:00 to 8:00 and shot in an area B from 17:00 to 19:00; personnel in the personnel group are frequently driven, and the affinity integral of the personnel in the personnel group and the personnel with r service attributes is higher than a threshold value), and the like. The multi-dimensional analysis of the personnel group can make the characteristic or rule which is not obvious in the single dimension appear to be obvious as the group characteristic. After the group characteristics are obtained, the group characteristics can be used as new archive labels, and a closed loop is formed with the scheme for determining the personnel identity attribute according to the archive labels, so that the personnel identity attribute determined based on the new archive labels and the previous archive labels is more accurate; and determining a target person group matched with the preset group characteristic from the plurality of person groups according to the preset group characteristic.

The basic attribute analysis in the multi-dimensional analysis is actually a basic attribute of a person of the statistical person group, and the archive tag analysis in the multi-dimensional analysis is actually an archive tag of a person of the statistical person group.

It should be noted that, the above-mentioned portrait archive set with personnel identity attribute information also supports manual batch addition of portrait archives or single addition of portrait archives (the personnel identity attribute of the added portrait archives matches with the group identity attribute information of the portrait archive set). In addition, along with the change of the corresponding relation between the file label and the personnel identity attribute, the obtained portrait file set and personnel group are updated.

The above description describes the process of determining the identity attribute of the person and the process of dividing the person group, and the screening process of the person group is described in detail below.

In the embodiment of the invention, after the portrait file collection is obtained, the method further comprises the following steps: performing group identification on personnel in the target personnel group corresponding to the target portrait archive set based on group discovery conditions to obtain personnel groups; the target image file set is a subset of the image file set, and the subset is an image file set matched with a preset group identity attribute and/or a preset group characteristic in the image file set.

The preset group identity attribute is a person identity attribute of interest or attention of the user, and the preset group characteristic is a person group characteristic of interest or attention of the user.

Specifically, referring to fig. 4, the step of performing group identification on the people in the target people group corresponding to the target portrait archive set based on the group discovery condition includes:

step S401, determining the intimacy between target persons in the target person group; wherein the target personnel are any two personnel in the target personnel group;

Step S402, if the affinity between the target persons is greater than a preset affinity threshold, determining that the target persons are relationship persons;

Step S403, building a relationship chain by taking a relational person as a node;

Step S404, determining a target relationship chain in the relationship chain; wherein the target relationship chain is any one of the relationship chains;

Step S405, if the number of nodes of the target relationship chain is greater than a preset threshold, using the relational personnel corresponding to the nodes of the target relationship chain as personnel in the personnel group, and further obtaining the personnel group.

And determining the intimacy between the target personnel according to the number of the same-line times and the same-line occasions of the target personnel. The calculation method of the number of the same line is not particularly limited, and the same line may be taken by the same camera once for the same line at the same time, or two persons may be taken by the same camera once for the same line within a certain time range (for example, within 15 s), or two persons may be taken by cameras within a certain distance range (for example, 10 m) simultaneously for the same line. Two siblings occurring within a certain time frame (e.g. half an hour) may also be considered as one sibling to avoid repetition of the counting. In one example, for example, target persons a and B are on the same line 15 times during the time of day, wherein the time interval between 5 same lines is more than half an hour, and the time interval between 10 same lines and the 5 same lines is less than half an hour, then the affinity between a and B is 5 (i.e. the affinity between a and B is divided into 5 points), and 10 same lines with the 5 same lines less than half an hour are not counted separately. In addition, different weights can be set for different peer occasions, for example, the weight of the peer arranged in the hotel is w1, the weight of the peer in the cell is w2, and the like, and when analysis results in that A and B are n times in the hotel peer and m times in the cell, the obtained intimacy is determined as follows: w 1x n+w2 x m. In determining the intimacy, there may be multiple implementation manners specifically, which are not limited by the embodiments of the present invention.

After the intimacy between the target persons is obtained, further judging whether the intimacy between the target persons is larger than a preset intimacy threshold, if so, determining that the target persons are relationship persons, and constructing a relationship chain by taking the relationship persons as nodes.

Therefore, a plurality of relation chains can be constructed, each item of relation chain in the relation chains is traversed, and if the number of nodes of the target relation chain is larger than a preset threshold, the relation personnel corresponding to the nodes of the target relation chain are used as personnel in personnel groups, so that the personnel groups are obtained. The embodiment of the invention does not limit the process of determining the personnel group specifically, and other implementation manners are also possible.

In the actual application process, along with the increase of the portrait snapshot, the previous intimacy of the target personnel in the target personnel group also changes in real time. The above group discovery conditions may be used to periodically identify groups of targeted persons in a targeted person population.

As the personnel in the personnel group are continuously updated and increased, the intimacy among the target personnel in the target personnel group is also continuously updated, and the member composition and relationship network of the personnel group obtained by determining are continuously approximate to the real world.

In addition, after the personnel group is obtained, group personnel can be manually added in the personnel group, so that the personnel group is closer to the real physical world.

In this embodiment, after the community of people is obtained, the method further comprises the following processes of (i) and (ii):

(i) Determining the intimacy between the personnel in the personnel group and the personnel outside the personnel group based on the personnel profile of the personnel in the personnel group and the personnel profile of the personnel outside the personnel group;

(ii) And determining the edge personnel of the personnel group according to the intimacy.

The edge personnel can be personnel in the personnel group or not, and the personnel is determined to be the edge personnel of the personnel group as long as the intimacy degree of the personnel in the personnel group is larger than a preset threshold value, and the process can excavate the personnel which are not in the group but are closely related to the group in the group, so that deep excavation and line expansion of the personnel group are realized.

In this embodiment, after the staff is obtained, the method further includes the following processes 1) and 2):

1) Carrying out multidimensional analysis on the portrait files corresponding to the personnel groups to obtain group characteristics of the personnel groups; wherein the multidimensional analysis comprises at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, place dimension analysis and affinity dimension analysis;

2) And determining the target personnel group in the personnel groups according to the preset group characteristics.

The multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, location dimension analysis, and affinity dimension analysis. Corresponding to the characteristic, the characteristic of the group comprises basic attribute information, archive labels, appearance time characteristics, location characteristics and identity attribute characteristics of staff in the staff group, (such as that the staff in the staff group is frequently shot in a C area of 7:00-9:00 and is shot in a D area of 18:00-20:00, the staff in the staff group is always in a motor vehicle, and the affinity score of the staff in the staff group and the staff with r identity attribute is higher than a threshold value), and the like.

After the community characteristics are obtained, a target community of people matched with the preset community characteristics can be determined from a plurality of communities of people according to the preset community characteristics. The preset community feature may be a community feature of people that are of interest or interest to the user.

According to the method for determining the personnel identity attribute, on the basis of determining the personnel identity attribute, the personnel are divided into different groups based on the identified personnel identity attribute, then the groups formed by the same type of personnel are subjected to research analysis, the target personnel group (namely suspected s group) is determined from the personnel groups, the closely-connected personnel groups (such as group partners in s group) are found in the target personnel group based on group finding conditions, and various personnel group research analysis means are provided, so that the efficiency of finding the personnel groups by related departments is greatly improved, and the automatic research analysis of big data is realized.

Example 3:

In accordance with an embodiment of the present invention, there is provided an embodiment of a community screening method, it being noted that the steps shown in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order other than that shown or described herein.

Fig. 5 is a flowchart of a community screening method according to an embodiment of the present invention, as shown in fig. 5, the method comprising the steps of:

Step S501, a portrait file of a person to be identified is obtained; the portrait file at least comprises: the method comprises the steps of taking a portrait snapshot, taking time of the portrait snapshot and information of an image acquisition device for taking the portrait snapshot, wherein each portrait file corresponds to a person to be identified;

step S502, determining archive labels and/or basic attribute information according to the portrait archive;

step S503, according to the corresponding relation between the file label and/or the basic attribute information and the personnel identity attribute, the personnel identity attribute of the personnel to be identified corresponding to the personnel file is determined;

Step S504, dividing a plurality of portrait files with personnel identity attribute determined to obtain a portrait file set with personnel identity attribute information;

Step S505, carrying out group identification on personnel in a target personnel group corresponding to the target portrait archive set based on group discovery conditions to obtain personnel groups; the target portrait archive set is a subset of the portrait archive set, and the subset is a portrait archive set matched with the preset group identity attribute and/or the preset group characteristic in the portrait archive set.

It should be noted that, in the steps related to the community screening method of embodiment 3, the steps related to the method for determining the identity attribute of the person of embodiment 2 are the same, and are not repeated, and the content described in embodiment 2 can be used in embodiment 3.

Example 4:

The embodiment of the invention also provides a device for determining the personnel identity attribute, which is mainly used for executing the method for determining the personnel identity attribute provided by the embodiment of the invention, and the device for determining the personnel identity attribute provided by the embodiment of the invention is specifically introduced.

Fig. 6 is a schematic diagram of a device for determining a person identity attribute according to an embodiment of the present invention, and as shown in fig. 6, the device for determining a person identity attribute mainly includes: an acquisition unit 10, a first determination unit 20, and a second determination unit 30, wherein:

the acquisition unit is used for acquiring the portrait file of the person to be identified; the portrait file at least comprises: the method comprises the steps of taking a portrait snapshot, taking time of the portrait snapshot and information of an image acquisition device for taking the portrait snapshot, wherein each portrait file corresponds to a person to be identified;

The first determining unit is used for determining archive labels and/or basic attribute information according to the portrait archive;

And the second determining unit is used for determining the personnel identity attribute of the personnel to be identified corresponding to the portrait file according to the corresponding relation between the archive label and/or the basic attribute information and the personnel identity attribute.

In the embodiment of the invention, a portrait file of a person to be identified is firstly obtained; then, determining archive labels and/or basic attribute information according to the portrait archive; and finally, determining the personnel identity attribute of the personnel to be identified corresponding to the personnel image file according to the corresponding relation between the file label and/or the basic attribute information and the personnel identity attribute. As can be seen from the above description, the method of the present invention can determine the personnel identity attribute of the personnel to be identified corresponding to the personnel profile based on the personnel profile, the profile label and/or the corresponding relation between the basic attribute information and the personnel identity attribute, and the process can determine the identity attribute of the personnel without the real photo library; in addition, the process can also obtain the identity information of the personnel to be identified and the business attribute information of the personnel to be identified according to the information of the personnel identity attribute in the corresponding relation, so that the technical problem that the identity attribute of the personnel is difficult to effectively determine in the prior art is solved.

Optionally, the archive label includes: at least one of an abnormal behavior type file tag, a personnel state type file tag, a situation type file tag and an affinity type file tag; the archive labels of the abnormal behavior categories include: daytime and night out labels, frequent out labels, personnel gathering labels and regional resident labels; the profile tag for the personnel status category includes: a resident tag, an leave district tag, and a visit area tag; the archive labels of the show and miss categories include: in the preset time, labels which do not reach the preset times are displayed in the same image acquisition device or the same area; the archive label of the affinity class includes: a label of the same-line times of the person with the known person identity attribute; the first determining unit is further configured to perform statistical analysis on the portrait file to obtain a file label and/or basic attribute information.

Optionally, the second determining unit is further configured to: determining a target portrait file in a plurality of portrait files according to preset basic attribute information; and matching the file label of the target portrait file with the file label corresponding to the preset basic attribute information in the corresponding relation, and determining the personnel identity attribute of the personnel to be identified corresponding to the target portrait file according to the matching result.

Optionally, the device is further configured to: dividing a plurality of portrait files with the personnel identity attribute determined to obtain a portrait file set with personnel identity attribute information; taking the personnel identity attribute information of the personnel image file set as the group identity attribute of the personnel group corresponding to the personnel image file set; and/or carrying out multidimensional analysis on the portrait files in the portrait file set to obtain group characteristics of a personnel group; the multidimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, location dimension analysis, and affinity dimension analysis.

Optionally, the device is further configured to: performing group identification on personnel in the target personnel group corresponding to the target portrait archive set based on group discovery conditions to obtain personnel groups; the target set of portrait files is a subset of the set of portrait files.

Optionally, the subset is a set of portrait files matching a preset group identity attribute and/or a preset group feature.

Optionally, the device is further configured to: determining the intimacy between target persons in a target person group; the target personnel are any two personnel in the target personnel group; if the affinity between the target persons is greater than a preset affinity threshold, determining that the target persons are relational persons; building a relationship chain by taking relational personnel as nodes; determining a target relationship chain in the relationship chain; the target relationship chain is any relationship chain in the relationship chains; if the number of the nodes of the target relation chain is larger than a preset threshold, taking the relational personnel corresponding to the nodes of the target relation chain as personnel in the personnel group, and further obtaining the personnel group.

Optionally, the device is further configured to: determining the intimacy between the personnel in the personnel group and the personnel outside the personnel group based on the personnel profile of the personnel in the personnel group and the personnel profile of the personnel outside the personnel group; and determining the edge personnel of the personnel group according to the intimacy.

Optionally, the device is further configured to: carrying out multidimensional analysis on the portrait files corresponding to the personnel groups to obtain group characteristics of the personnel groups; the multidimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, place dimension analysis and affinity dimension analysis; and determining the target personnel group in the personnel groups according to the preset group characteristics.

The device provided by the embodiment of the present invention has the same implementation principle and technical effects as those of the foregoing method embodiment, and for the sake of brevity, reference may be made to the corresponding content in the foregoing method embodiment where the device embodiment is not mentioned.

In another implementation of the present invention, there is also provided a computer storage medium having stored thereon a computer program which, when run, performs the steps of the method of any of the above method embodiments 2.

In addition, in the description of embodiments of the present invention, unless explicitly stated and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.

In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.

In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

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

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random AccessMemory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for determining a person identity attribute, applied to a person analysis system, comprising:

acquiring a portrait file of a person to be identified; the portrait file at least comprises: the system comprises a portrait snapshot, snapshot time of the portrait snapshot and information of an image acquisition device for snapshot of the portrait snapshot, wherein each portrait file corresponds to a person to be identified;

determining archive labels and/or basic attribute information according to the portrait archive;

According to the corresponding relation between the archive label and/or the basic attribute information and the personnel identity attribute, the personnel identity attribute of the personnel to be identified corresponding to the portrait archive is determined;

after obtaining the personnel identity attribute of the personnel to be identified corresponding to the portrait file, the method further comprises:

Dividing a plurality of portrait files with the personnel identity attribute determined to obtain a portrait file set with personnel identity attribute information;

Taking the personnel identity attribute information of the personnel image file set as the group identity attribute of the personnel group corresponding to the personnel image file set; and/or carrying out multidimensional analysis on the portrait files in the portrait file collection to obtain group characteristics of the personnel group, and taking the group characteristics as new file labels;

the multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, place dimension analysis and affinity dimension analysis;

the method further comprises the steps of:

Performing group identification on personnel in the target personnel group corresponding to the target portrait archive set based on group discovery conditions to obtain personnel groups; the target portrait archive set is a subset of the portrait archive set;

Determining the intimacy between the personnel in the personnel group and the personnel outside the personnel group based on the personnel profile of the personnel in the personnel group and the personnel outside the personnel group;

And determining the edge personnel of the personnel group according to the intimacy.

2. The method of claim 1, wherein the archive label comprises: at least one of an abnormal behavior type file tag, a personnel state type file tag, a situation type file tag and an affinity type file tag;

The archive tag of the abnormal behavior category comprises: daytime and night out labels, frequent out labels, personnel gathering labels and regional resident labels;

The profile tag of the personnel status category includes: a resident tag, an leave district tag, and a visit area tag;

The archive label of the present situation category comprises: in the preset time, labels which do not reach the preset times are displayed in the same image acquisition device or the same area;

the archive label of the affinity class comprises: a label of the same-line times of the person with the known person identity attribute;

And carrying out statistical analysis on the portrait file to obtain the file label and/or the basic attribute information.

3. Method according to claim 1 or 2, wherein the step of determining the person identity attribute of the person to be identified corresponding to the person profile according to the profile tag and/or the correspondence of the base attribute information and the person identity attribute comprises:

determining a target portrait file in a plurality of portrait files according to preset basic attribute information;

And matching the archive label of the target portrait archive with the archive label corresponding to the preset basic attribute information in the corresponding relation, and determining the personnel identity attribute of the personnel to be identified corresponding to the target portrait archive according to the matching result.

4. The method of claim 1, wherein the subset is a set of portrait files in the set of portrait files that match a preset group identity attribute and/or a preset group feature.

5. The method of claim 1, wherein the step of community identification of persons in the target person group corresponding to the target person profile set based on community finding conditions comprises:

Determining the intimacy between target persons in the target person group; the target personnel are any two personnel in the target personnel group;

if the affinity between the target persons is greater than a preset affinity threshold, determining that the target persons are relational persons;

building a relationship chain by taking the relational personnel as nodes;

determining a target relationship chain among the relationship chains; the target relationship chain is any one of the relationship chains;

And if the number of the nodes of the target relation chain is larger than a preset threshold, taking the relational personnel corresponding to the nodes of the target relation chain as personnel in the personnel group, and further obtaining the personnel group.

6. The method of claim 5, wherein the method further comprises:

carrying out multidimensional analysis on the portrait files corresponding to the personnel groups to obtain group characteristics of the personnel groups; the multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, place dimension analysis and affinity dimension analysis;

And determining a target personnel group in the personnel groups according to the preset group characteristics.

7. A device for determining identity attributes of a person, the device being applied to a person analysis system and comprising:

The acquisition unit is used for acquiring the portrait file of the person to be identified; the portrait file at least comprises: the system comprises a portrait snapshot, snapshot time of the portrait snapshot and information of an image acquisition device for snapshot of the portrait snapshot, wherein each portrait file corresponds to a person to be identified;

the second determining unit is used for determining the personnel identity attribute of the personnel to be identified corresponding to the portrait file according to the corresponding relation between the file label and/or the basic attribute information and the personnel identity attribute;

The device is also for: dividing a plurality of portrait files with the personnel identity attribute determined to obtain a portrait file set with personnel identity attribute information; taking the personnel identity attribute information of the personnel image file set as the group identity attribute of the personnel group corresponding to the personnel image file set; and/or carrying out multidimensional analysis on the portrait files in the portrait file collection to obtain group characteristics of the personnel group, and taking the group characteristics as new file labels; the multi-dimensional analysis includes at least one of: basic attribute analysis, archive tag analysis, time dimension analysis, place dimension analysis and affinity dimension analysis;

the device is also for: performing group identification on personnel in the target personnel group corresponding to the target portrait archive set based on group discovery conditions to obtain personnel groups; the target portrait archive set is a subset of the portrait archive set; determining the intimacy between the personnel in the personnel group and the personnel outside the personnel group based on the personnel profile of the personnel in the personnel group and the personnel outside the personnel group; and determining the edge personnel of the personnel group according to the intimacy.

8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of the preceding claims 1 to 6 when executing the computer program.

9. A computer storage medium, characterized in that a computer program is stored thereon, which computer program, when running, performs the steps of the method according to any of the preceding claims 1 to 6.