WO2022257306A1

WO2022257306A1 - Identity identification method and apparatus, electronic device, and storage medium

Info

Publication number: WO2022257306A1
Application number: PCT/CN2021/122138
Authority: WO
Inventors: 郑莞蓉; 蒋小可
Original assignee: 成都商汤科技有限公司
Priority date: 2021-06-11
Filing date: 2021-09-30
Publication date: 2022-12-15
Also published as: CN113269129A

Abstract

The present invention relates to an identity identification method and apparatus, an electronic device, and a storage medium. The method comprises: obtaining a collected image and temporal-spatial information of a first user, wherein the temporal-spatial information is used for indicating collection time and a collection location of the collected image; determining at least one second user in a user database according to the collected image of the first user; respectively determining historical temporal-spatial distribution of the at least one second user, wherein the historical temporal-spatial distribution is used for representing a probability that the second user appears at one or more locations in at least one time period; and identifying the first user in the at least one second user on the basis of the temporal-spatial information of the first user and the historical temporal-spatial distribution of the at least one second user. Embodiments of the present invention can improve the accuracy rate of identity identification.

Description

Identification method and device, electronic device and storage medium

This application claims the priority of the Chinese patent application filed on June 11, 2021, with the application number 202110654499.8, and the title of the invention is "identification method and device, electronic equipment and storage medium", the entire content of which is incorporated in this application by reference .

technical field

The present disclosure relates to the field of computer technology, and in particular to an identification method and device, electronic equipment and a storage medium.

Background technique

In recent years, face recognition technology has developed rapidly in various industries. For example, in the field of rail transit, face recognition technology can be used as a way to sell tickets. By adding face recognition-based ticketing or ticket checking equipment in rail transit, ticket sales or ticket checking can be realized by collecting facial images of passengers. This method can increase the speed of ticket sales or ticket inspection and reduce unnecessary congestion during peak hours.

Contents of the invention

The present disclosure proposes an identification technical solution.

According to an aspect of the present disclosure, there is provided an identification method, including: acquiring a first user's captured image and spatio-temporal information, wherein the spatio-temporal information is used to indicate the time and location of the captured image; according to the The collected images of the first user, determine at least one second user in the user database; respectively determine the historical spatio-temporal distribution of the at least one second user, wherein the historical spatio-temporal distribution is used to indicate that the second user is at least Probability of a time period occurring at one or more locations; identifying the first user among the at least one second user based on the spatiotemporal information of the first user and the historical spatiotemporal distribution of the at least one second user .

In one or more possible implementations, the identifying the first user among the at least one second user based on the spatiotemporal information of the first user and the historical spatiotemporal distribution of the at least one second user , comprising: based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, determining the spatio-temporal similarities between the first user and the at least one second user; identifying the first user among the at least one second user based on the spatio-temporal similarity between the user and the at least one second user; or, based on the first user's relationship with the at least one second user The visual similarity and the spatio-temporal similarity are used to identify the first user among the at least one second user.

In one or more possible implementations, based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, determining the relationship between the first user and the at least one second user The spatio-temporal similarity includes: for at least one second user, in the historical spatio-temporal distribution of the second user, determining the probability that the second user is located at the collection location at the collection time; according to the A probability determines a spatio-temporal similarity of the first user to the second user.

In one or more possible implementations, the historical spatiotemporal distribution includes an inbound distribution and an outbound distribution, and in the historical spatiotemporal distribution of the second user, it is determined that the second user is The probability of being located at the collection location includes: when the collection location is an outbound station, in the outbound distribution of the second user, determining the time period corresponding to the collection time of the second user The probability of being located at the outbound site; or, in the case where the collection site is an inbound site, in the inbound distribution of the second user, determine the time corresponding to the collection time of the second user The probability that a segment is located at the inbound site.

In one or more possible implementation manners, the identifying the at least one second user among the at least one second user based on the visual similarity and the spatiotemporal similarity between the first user and the at least one second user respectively The first user includes: for at least one second user, weighting the visual similarity and spatio-temporal similarity between the first user and the second user to obtain the first user and the second user Fusion similarity of users: identifying the first user among the at least one second user based on the fusion similarities between the first user and the at least one second user respectively.

In one or more possible implementation manners, the identifying the first user among the at least one second user based on the fusion similarities between the first user and the at least one second user respectively includes : Determining, among the at least one second user, a second user whose fusion similarity with the first user is the largest, as the second user matching the first user.

In one or more possible implementation manners, the determining at least one second user in the user database according to the captured image of the first user includes: combining the captured image of the first user with the user database Match the user images of historical users in the database to determine the visual similarity between the first user and the historical user; if the visual similarity is greater than a preset threshold, determine the historical user as the second user.

In one or more possible implementations, the method further includes: acquiring historical user records of multiple historical users, wherein the multiple historical users include the at least one second user; based on the historical user records , counting the historical number of times that the historical user appeared in one or more places in at least one time period, and counting the total number of times that one or more historical users appeared in the place in the time period; according to the historical times and The total number of times generates the historical spatiotemporal distribution of the historical users.

In one or more possible implementation manners, the method further includes: smoothing the historical spatio-temporal distribution to obtain the historical spatio-temporal distribution of the historical users after the smoothing processing.

In one or more possible implementation manners, after identifying the first user among the at least one second user, further includes: saving the captured image and spatio-temporal information of the first user in the second user A historical user record of a user, wherein the historical user record includes at least one of the following information: user ID, inbound ID or outbound ID, site ID, time of inbound or outbound.

According to an aspect of the present disclosure, an identification device is provided, including:

An acquisition module, configured to acquire the collected image and spatio-temporal information of the first user, wherein the spatio-temporal information is used to indicate the collection time and location of the collected image;

A first determining module, configured to determine at least one second user in the user database according to the captured image of the first user;

A second determining module, configured to respectively determine the historical spatiotemporal distribution of the at least one second user, wherein the historical spatiotemporal distribution is used to represent the probability that the second user appears in one or more places in at least one time period ;

An identifying module, configured to identify the first user among the at least one second user based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user.

In one or more possible implementation manners, the identification module is configured to determine, based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, the a spatio-temporal similarity of at least one second user; identifying the first user among the at least one second users based on the spatio-temporal similarities between the first user and the at least one second user, respectively; or, Identifying the first user among the at least one second users based on the visual similarity and the spatio-temporal similarity between the first user and the at least one second user respectively.

In one or more possible implementations, the identification module is configured to, for at least one of the second users, determine that the second user is in the collection time The probability of being located at the collection location; determining the spatio-temporal similarity between the first user and the second user according to the probability.

In one or more possible implementations, the identification module is configured to determine, in the outbound distribution of the second user, that the second user is The probability that the time period corresponding to the collection time is located at the outbound site; or, in the case that the collection site is an inbound site, in the inbound distribution of the second user, determine the second user The probability of being at the inbound site during the time period corresponding to the collection time.

In one or more possible implementation manners, the recognition module is configured to, for the at least one second user, weight the visual similarity and spatiotemporal similarity between the first user and the second user, Obtain the fusion similarity between the first user and the second user; identify the first user among the at least one second user based on the fusion similarity between the first user and the at least one second user respectively a user.

In one or more possible implementation manners, the identification module is configured to determine, among the at least one second user, the second user whose fusion similarity with the first user is the largest as the first user Matched second user.

In one or more possible implementations, the identification module is configured to match the captured image of the first user with user images of historical users in the user database, and determine whether the first user is related to the The visual similarity of the historical user; if the visual similarity is greater than a preset threshold, the historical user is determined as the second user.

In one or more possible implementations, the device further includes: a generating module, configured to acquire historical user records of multiple historical users, where the multiple historical users include the at least one second user; based on The historical user record counts the historical number of times that the historical user appears in one or more places in at least one time period, and counts the total number of times that one or more historical users appear in the place in the time period; according to The historical times and the total times generate the historical spatio-temporal distribution of the historical users.

In one or more possible implementation manners, the generation module is further configured to perform smoothing processing on the historical spatio-temporal distribution, to obtain the historical spatio-temporal distribution of the historical users after smoothing processing.

In one or more possible implementations, the device further includes: a recording module, configured to save the captured image and spatio-temporal information of the first user in the historical user record of the first user, wherein the The historical user record includes at least one of the following information: user ID, inbound ID or outbound ID, site ID, inbound or outbound time.

According to an aspect of the present disclosure, there is provided an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above-mentioned method.

According to one aspect of the present disclosure, there is provided a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above method is implemented.

According to an aspect of the present disclosure, there is provided a computer program product, including computer readable codes, or a non-volatile computer readable storage medium bearing computer readable codes, when the computer readable codes are stored in an electronic device When running in the processor, the processor in the electronic device is used to implement the above method.

In the embodiment of the present disclosure, the captured image and spatio-temporal information of the first user can be obtained, and at least one second user can be determined in the user database based on the captured image of the first user, and then the historical spatio-temporal information of at least one second user can be respectively determined. distribution, the historical spatio-temporal distribution can represent the probability that the second user appears in one or more places in at least one time period, so that based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of at least one second user, in at least one second The first user is identified among the users. In this way, when confirming the user identity of the first user, the user image of the first user can be combined with spatio-temporal information as the basis for identification, which improves the accuracy of identification and reduces the risk of identification due to the use of facial images alone. resulting in misjudgment.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings.

Description of drawings

The accompanying drawings here are incorporated into the description and constitute a part of the present description. These drawings show embodiments consistent with the present disclosure, and are used together with the description to explain the technical solution of the present disclosure.

Fig. 1 shows a flowchart of an identification method according to an embodiment of the present disclosure.

FIG. 2 shows a distribution diagram of historical spatiotemporal distribution according to an embodiment of the disclosure.

FIG. 3 shows a distribution diagram of an unsmoothed historical spatiotemporal distribution provided by an embodiment of the present disclosure.

Fig. 4 shows a flowchart of an example of an identification method according to an embodiment of the present disclosure.

Fig. 5 shows a block diagram of an identity recognition device according to an embodiment of the present disclosure.

Fig. 6 shows a block diagram of an electronic device according to an embodiment of the present disclosure.

Fig. 7 shows a block diagram of an electronic device according to an embodiment of the present disclosure.

Detailed ways

Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numbers in the figures indicate functionally identical or similar elements. While various aspects of the embodiments are shown in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as superior or better than other embodiments.

The term "and/or" in this article is just an association relationship describing associated objects, which means that there can be three relationships, for example, A and/or B can mean: A exists alone, A and B exist simultaneously, and there exists alone B these three situations. In addition, the term "at least one" herein means any one of a variety or any combination of at least two of the more, for example, including at least one of A, B, and C, which may mean including from A, Any one or more elements selected from the set formed by B and C.

In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific implementation manners. It will be understood by those skilled in the art that the present disclosure may be practiced without some of the specific details. In some instances, methods, means, components and circuits that are well known to those skilled in the art have not been described in detail so as to obscure the gist of the present disclosure.

For urban rail transit scenarios, when face recognition-based ticket sales and inspection methods are used to identify current passengers, there may be multiple user images that are very similar to the face images of current passengers in terms of visual features, making it difficult to judge The user identity of the current passenger.

The identification scheme provided by the embodiments of the present disclosure can be applied to indoor and outdoor scenes such as rail transit, urban subway, scenic spots, and tourist exhibition halls. For example, in a large and medium-sized urban rail transit with a large passenger flow, the identification scheme provided by the embodiment of the present disclosure can be used to verify the identity of passengers entering and leaving the station. In addition to using traditional face recognition technology to authenticate users In addition, it is also possible to combine the spatio-temporal information of passengers entering and exiting the station into the process of face recognition, and assist in identifying passengers through spatio-temporal information. For another example, in a scenic spot with multiple scenic spots, the identification scheme provided by the embodiments of the present disclosure can verify the identity of tourists entering and leaving the scenic spot, and the spatiotemporal information of tourists entering and leaving the scenic spot can assist in identifying the identity of tourists. In this way, the phenomenon of misjudgment caused by simply using face images for identity recognition due to the increase in user traffic can be reduced to a large extent, thereby improving the accuracy of identity recognition.

In the embodiment of the present disclosure, the identification method may be performed by electronic devices such as terminal equipment or servers, and the terminal equipment may be user equipment (User Equipment, UE), mobile equipment, user terminal, cellular phone, cordless phone, personal digital assistant (Personal Digital Assistant, PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc., the method can be implemented by calling the computer-readable instructions stored in the memory by the processor. Alternatively, the method may be performed by a server. The identity recognition method in the embodiment of the present disclosure will be described below by taking an electronic device as an execution subject as an example.

Fig. 1 shows a flowchart of an identification method according to an embodiment of the present disclosure. As shown in Fig. 1, the identification method includes:

Step S11, acquiring the collected image and spatio-temporal information of the first user.

In this embodiment of the present disclosure, the first user may be a user entering or leaving the target area, and the collected image may be an image captured for the first user. Spatio-temporal information can be used to indicate the collection time and collection location of the collected images. The instant-spatial information can include the collection time and collection location. The collection time and collection location of the collected images can be considered as the corresponding time and time when the first user enters or leaves the target area. Place. The target area may include multiple locations. For example, the target area may be an underground station, a rail transit station, a scenic spot, etc., and the locations included in the target area may be exits or entrances set in the target area. The first user can enter or leave the target area at any location. For example, when the target area is a subway station, the first user can enter or exit at any station of the subway station.

In some implementation manners, the electronic device may have a photographing function, and may photograph the target area in real time to obtain the captured image and spatio-temporal information of the first user entering the target area. In some implementation manners, a camera device may be installed in the target area, and the target area may be photographed in real time, and the electronic device may acquire images and spatio-temporal information of the first user of the camera device in a wired or wireless manner.

Step S12, determining at least one second user in the user database according to the captured image of the first user.

In an embodiment of the present disclosure, the captured image of the first user may be matched with user images of multiple historical users in the user database, and at least one second user matching the first user may be determined in the user database. The second user may be a historical user who visually matches the first user. User images of multiple historical users may be stored in the user database, each historical user may correspond to one or more user images, and the user images may include face images and/or body images.

In some implementations, the captured image of the first user can be matched with the user images of the historical users in the user database to determine the visual similarity between the first user and the historical users, for example, the person of the first user in the captured image can be extracted Face features, then calculate the distance (such as Euclidean distance) between the face features of the first user and the face features of at least one user image in the user database, and obtain the visual similarity between the collected images and the user images of multiple historical users in the database Spend. The visual similarity can be regarded as the visual similarity between the first user and the historical users. Further, the visual similarity between the captured image and at least one user image may be compared with a preset threshold, and if the visual similarity is greater than the preset threshold, the historical user corresponding to the corresponding user image may be determined as the second user. In this way, at least one second user matching the first user can be preliminarily determined among multiple historical users.

In some implementations, when the face of the first user is occluded or difficult to recognize, the human body features of the first user in the captured image may also be extracted, and then the human body features of the first user may be combined with the multiple data in the user database. The human body features of each user image are matched to determine the visual similarity between the first user and multiple historical users.

In some implementations, in order to reduce the amount of calculation for image matching, user images of multiple historical users in the user database may be classified and stored according to visual similarity, and the visual similarity between user images corresponding to the same category is relatively high. When matching the captured image of the first user with the user images of multiple historical users in the user database, the facial features or human body features of the captured image of the first user can be compared with the central features of one or more classes Yes, determine the class that matches the first user, and then match the captured image of the first user with user images of one or more historical users in the class, and determine at least one second user that matches the first user. Wherein, the central feature of a class may be an average or median value of face features or human body features of at least one user image in the class.

In some implementations, according to the visual similarities between the first user and multiple historical users in the database, when at least one second user is determined in the user database, the visual similarity between the first user and multiple historical users can also be The plurality of historical users are sorted in descending order, and then a preset number of historical users sorted first are determined as the second users. For example, the top 5 sorted historical users are determined as the second users.

Step S13, respectively determining the historical spatio-temporal distribution of the at least one second user.

In the embodiments of the present disclosure, the historical spatio-temporal distribution is used to represent the probability that the second user appears in one or more places in at least one time period. Different historical users may correspond to different historical spatiotemporal distributions. The probability of one or more historical users appearing in one or more places can be determined through the historical spatio-temporal distribution. For example, according to the historical spatio-temporal distribution of the second user, the time of the second user at 9:00-10:00 am can be determined The probability that a segment appears at location A is 70%. Historical spatio-temporal distribution can be generated based on historical user records. Historical user records can record the time and place when historical users entered or left the target area. Historical user records can provide guidance for current user spatio-temporal distribution, that is, a user in If there are more times of appearance at location B from 9:00 to 10:00, it is very likely that the current user appears at location B from 9:00 to 10:00.

Step S14, identifying the first user among the at least one second user based on the spatiotemporal information of the first user and the historical spatiotemporal distribution of the at least one second user.

In an embodiment of the present disclosure, based on the spatiotemporal information of the first user and the historical spatiotemporal distribution of at least one second user, the spatiotemporal similarity between the first user and at least one second user can be determined, for example, according to the spatiotemporal Information, in the historical spatio-temporal distribution of at least one second user, query the value corresponding to the spatio-temporal information of the first user, and determine the spatio-temporal similarity between the first user and the at least one second user according to the value. The spatio-temporal similarity can be understood as the degree of matching between the first user and the second user in time and place, that is, the possibility that the second user appears in the spatio-temporal information of the first user. The historical spatio-temporal distribution can be expressed in the form of statistical tables, statistical matrices or statistical curves.

In some implementations, after determining the spatio-temporal similarities between the first user and at least one second user, it is further possible to identify among at least one second user based on the spatio-temporal similarities between the first user and at least one second user. The first user, for example, may determine the second user having the greatest spatio-temporal similarity with the first user among at least one second user as the first user, and may further acquire identity information of the first user. In this way, the second user who is visually similar to the first user can be screened out first based on the collected images of the first user, and then based on the spatiotemporal similarity between the first user and at least one second user, at least one second user In order to identify the first user, in the case that it is difficult to identify the user's identity through the face image, the user's identity can be verified through the user's historical spatiotemporal distribution, and the accuracy of identity recognition can be improved.

In some implementations, after the first user is identified from multiple second users, the captured images and spatio-temporal information of the first user may also be saved in the historical user records of the first user, so that the collected images of the first user Images and spatio-temporal information can be used as historical user records for subsequent historical spatio-temporal distribution updates to facilitate subsequent information quantification statistics and calculations. For example, in the scene of subway entry and exit, the newly added entry and exit information can be saved in the passenger's ride record .

In some application scenarios, for example, in scenarios such as ground falls, high-speed rail entry and exit, or scenic spot tours, after the first user is identified from multiple second users, the first user's entry information and exit information can be used to identify the first user. The consumption information of the target user is generated through station information, so that the deduction of fees can be automatically realized according to the consumption information of the first user, the efficiency of ticket sales or ticket inspection can be improved, and the safety and accuracy of automatic ticket sales or ticket inspection can be improved.

The embodiments of the present disclosure can consider the user's historical spatiotemporal distribution in the user's identity recognition, provide reference for the current identity recognition through the historical space-time distribution, and reduce the occurrence of misjudgment due to purely using face images for identity recognition. In scenarios such as rail transit and scenic spots with a large passenger flow and a large number of face images, it is not a small challenge to simply use face recognition technology to find the most similar face among millions of face images. In some cases where the face is partially blocked by wearing a mask, glasses, etc., and it is difficult to confirm the user's identity only through face recognition, the user's identity can be verified by assisting a single face recognition task through the user's historical spatiotemporal distribution, improving Accuracy of identification.

In some implementation manners, the first user may be identified among the at least one second user based on the visual similarity and the spatio-temporal similarity between the first user and the at least one second user. For example, the visual similarity and spatio-temporal similarity between the first user and at least one second user can be added to generate a score. The higher the score, the higher the possibility that the second user is the first user. Scores of a user and at least one second user are sorted from large to small, and then the second user with the highest score in the sorted sequence is determined as the target user. In this way, the visual similarity and spatio-temporal similarity between the first user and the second user can be combined to further improve the accuracy of identification.

In some implementations, when identifying the first user among at least one second user based on the visual similarity and spatiotemporal similarity between the first user and at least one second user, the at least one second user may also be , weighting the visual similarity and spatio-temporal similarity between the first user and at least one second user to obtain the fused similarity between the first user and at least one second user. Then, based on the fused similarities between the first user and the at least one second user, the first user is identified among the at least one second user. For example, the second user whose fusion similarity is greater than a certain similarity threshold can be determined as the first user, or the fusion similarities between multiple second users and the first user can be sorted from large to small, and then the Among the at least one second user, the second user whose fusion similarity with the first user is the largest is determined as the second user matching the first user. In this way, the target user can be identified through the fused similarity obtained by fusing the spatio-temporal similarity and the visual similarity, and the first user can be identified among multiple second users.

Here, the weight coefficients corresponding to the spatio-temporal similarity and the visual similarity can be preset, and then the spatio-temporal similarity and the visual similarity can be weighted and summed according to the preset weight coefficients to obtain the first user and at least one second user. Fusion Similarity. The weight of the spatiotemporal similarity and the weight coefficient of the visual similarity can be set according to actual needs. In some examples, the weight coefficient of the spatiotemporal similarity can be smaller than the weight coefficient of the visual similarity, so that the historical spatiotemporal distribution can be used as an auxiliary judgment for identification Information, using the historical spatiotemporal distribution to correct the misidentification caused by a single face image to judge the user's identity, and improve the accuracy of user identity recognition.

In some implementations, for at least one second user among the at least one second user, in the historical spatiotemporal distribution of the second user, it may be determined that the second user is located at the location corresponding to the first user at the collection time corresponding to the first user. The probability of the collection location, according to the determined probability, the spatio-temporal similarity between the first user and the second user can be determined.

Here, for the first user entering or leaving the target area, the known information of the first user may be the spatio-temporal information of the first user, and the spatio-temporal information may include collection time and collection location. For at least one second user, the historical spatiotemporal distribution of the second user may be determined according to the user identifier of the at least one second user. The historical spatio-temporal distribution may be represented by a distribution diagram or a table, and the present disclosure does not limit the specific representation. Then, in the historical spatio-temporal distribution of at least one second user, a value corresponding to the collection time and the collection location corresponding to the first user can be found, and the value can be the probability that the second user appears at the collection location at the collection time. Further, the probability may be determined as the spatio-temporal similarity between the first user and the second user. Alternatively, the spatiotemporal similarity corresponding to the probability can be determined according to the corresponding relationship between the probability and the spatiotemporal similarity. For example, the probability and the spatiotemporal similarity can have a linear relationship, and the spatiotemporal similarity can be obtained by multiplying the probability by a preset coefficient. In this way, the possibility of the second user matching the spatio-temporal information of the first user can be inferred by looking up the historical spatio-temporal distribution of the second user, that is, the spatio-temporal similarity between the first user and at least one second user can be determined, providing a user identification valid basis.

In some implementations, when determining the probability that the second user is located at the collection location corresponding to the first user at the collection time corresponding to the first user, the time period in which the collection time corresponding to the first user is located may be determined, and further in the second user From the historical spatio-temporal distribution of , determine the probability that the second user appears at the collection location during the time period. In this way, even if there are some errors in the collection time, by determining the time period where the collection time is located, the probability of the second user appearing at the collection location during this time period can be accurately determined, reducing the influence of possible errors in the collection time on the determined probability. influences.

FIG. 2 shows a distribution diagram of a historical spatio-temporal distribution according to an embodiment of the present disclosure. The historical spatio-temporal distribution may be a historical spatio-temporal distribution of a second user entering or leaving a target area at a location. The abscissa of the distribution graph can represent time, and the ordinate can represent probability. The ordinate can be between 0 and 1. In order to reduce the time scale marked on the abscissa, the duration can be quantified with a fixed time unit, for example, 30 minutes is used as the quantization unit of the time scale, that is, it can be understood as multiplying the time scale marked on the abscissa by 30 minutes The obtained time length may be equal to the actual time length. For example, in the subway scene, the user's ride time can be subtracted from the start time of the subway operation by the closing time of the subway operation, and then every 30 minutes can be regarded as a time period to obtain the abscissa of the distribution map. Assuming that the subway is closed The time is 18:00, the start time of the subway is 8:00, and the operation time of the subway is 10 hours in total. Taking 30 minutes as a time period, the operation time of the subway can be divided into 0-20 time periods. The abscissa in the figure is 0.0 It can represent the start time of 8:00, 2.5 can represent 9:15, and 20.0 can represent 18:00.

In some examples, the historical spatio-temporal distribution may include an inbound distribution and an outbound distribution. For example, in some rail transit scenarios, one or more historical users may correspond to an inbound distribution and an outbound distribution. The inbound distribution is different from the outbound distribution. When determining the probability that the second user is located at the collection location during the time period corresponding to the collection time, when the collection location of the first user is an outbound site, that is, in the outbound scenario, the outbound distribution of the second user can be , determine the probability that the second user is located at the outbound site in the time period corresponding to the collection time. Or, in the case where the first user's collection location is the inbound site, that is, in the inbound scenario, it can be determined that the second user is located in the inbound site during the time period corresponding to the collection time in the inbound distribution of the second user The probability.

For example, for the first user entering or exiting the station, the known quantity is whether the first user entered or exited the station, the current station and the collection time, according to the historical spatiotemporal distribution obtained from the historical user records , among multiple second users visually most similar to the first user, determine at least one second The probability that the user is the first user. If the historical spatio-temporal distribution shows that a second user has a higher probability of taking a bus at the current station and the time period where the collection time is located, then for this ride behavior, the second user is the same as the first user who is currently entering or leaving the station People are very likely, therefore, the basis for user identification can be provided through historical spatio-temporal distribution.

In an embodiment of the present disclosure, the spatio-temporal similarity between the first user and the at least one second user may be determined according to the historical spatio-temporal distribution of the at least one second user, thereby providing effective reference information for user identification. Among them, the historical spatio-temporal distribution can be generated based on the historical user records of historical users entering or leaving the target area. The historical user records can provide a reference for identifying the identity of the current first user. The historical spatio-temporal distribution is obtained through one or more implementation methods below process is described.

In some implementations, the historical user records of multiple historical users can be obtained, that is, the records of multiple historical users entering or leaving the target area can be obtained. The plurality of historical users may include at least one second user. Based on the historical user records of one or more historical users, it is possible to count the historical times of one or more historical users appearing in one or more places in at least one time period, and to count the historical times in which one or more historical users appeared in the time period Total counts for this location. Further, the historical spatiotemporal distribution of the historical users can be generated according to at least one time period, historical times and total times of one or more locations.

In this disclosure example, historical user records may represent the entry and exit records of historical users entering or leaving the target area. In order to make the historical spatio-temporal distribution credible, it is possible to obtain a recent period (not less than the preset duration) or multiple historical user records of historical users, for example, obtain the historical user records of one or more historical users in the last two weeks. The historical user record may include information such as the location where the historical user entered or left the target area, the time when the historical user entered or left the target area, and user identification. In order to facilitate the distinction between different locations, unique numbers can be assigned to different locations for identification. For the convenience of statistics, the time may be quantified in a certain time unit, for example, the time is quantified by taking thirty minutes as a time unit to obtain multiple time periods. According to the historical user records of one or more historical users, the historical times of one or more historical users appearing in one or more locations and at least one time period can be counted, so as to obtain a statistical model based on historical users, locations and time periods (such as statistical table, statistical matrix, etc.), the size of the statistical model can be: the number of historical users × the number of locations × the number of time periods. An element in the statistical model may represent the historical number of times the historical user entered or left the target area during the current time period at the current location. For one or more historical users, it is possible to count the historical times that the historical user appeared in one or more places in at least one time period, and the total number of times the historical users used in the statistics appeared in the place in the time period, each The historical times corresponding to the historical users are divided by the total times corresponding to all historical users, and the probability of the historical users appearing at the place in the time period can be obtained.

Here, a statistical model of one or more historical users appearing in one or more places in at least one time period can be generated by means of data modeling, and one or more historical users can be obtained in one or more location and at least one period of time entering or leaving the target area, that is, the current user's behavior of entering or leaving the target area can be predicted through historical user records.

In some examples, in the subway station riding scene, the historical user records at least one of the following information: user ID, entering or exiting identification, station identification, time of entering or exiting the station. The aforementioned statistical models may include inbound models and outbound models.

In some implementation manners, in order to further improve the reliability of the historical spatio-temporal distribution, the historical spatio-temporal distribution may be smoothed to obtain the historical spatio-temporal distribution of the historical users after the smoothing process. For example, Gaussian smoothing can be performed on the historical spatio-temporal distribution to obtain the smoothed historical spatio-temporal distribution. Since the probability values of adjacent time periods are continuous, for example, if a user frequently rides in a certain time period, there is also a certain probability of riding in an adjacent time period, so that Gaussian smoothing can be performed in the time dimension, making the historical space-time The distribution curve is also continuous in time. During Gaussian smoothing, the probability can be normalized, and the probability can be normalized between 0 and 1 to obtain the final historical spatiotemporal distribution of one or more historical users. Each value on the probability distribution curve of the historical space-time distribution may correspond to the probability that the historical user appears at the current location in the current time period.

FIG. 3 shows a distribution diagram of an unsmoothed historical spatio-temporal distribution provided by an embodiment of the present disclosure, and the foregoing FIG. 2 may be a distribution diagram of a smoothed historical spatio-temporal distribution. It can be seen that the probability curve corresponding to the smoothed historical space-time distribution has continuity, which can better meet the needs of actual scenarios.

The identification method provided by the embodiment of the present disclosure will be illustrated below through an example. Fig. 4 shows a flow chart of an example of an identification method according to an embodiment of the present disclosure. In this example, the target area may be an underground station, the first user may enter a subway station to take a passenger, and the captured image may be an entry into the station. For images, the collection location can be the station of the station, and the collection time can be the time of the station. As shown in Figure 4, the identification process provided by this example may include the following steps:

In step S201, the first user's inbound image, inbound time and inbound site are obtained.

Step S202, matching the inbound image of the first user with the user images of the historical users in the user database, and determining the visual similarity between the first user and the historical users.

Step S203, determining at least one second user whose visual similarity with the first user is greater than a preset threshold.

Step S204, acquiring the historical spatio-temporal distribution of at least one second user respectively.

Step S205, according to the entry time and entry site of the first user, respectively search the historical spatiotemporal distribution of at least one second user, and determine the spatiotemporal similarity between the first user and at least one second user;

Step S206, performing weighted fusion of the visual similarity and spatio-temporal similarity between the first user and at least one second user to obtain the fused similarity between the first user and at least one second user;

Step S207, sorting the fusion similarity between the first user and at least one second user from large to small, obtain the second user with the highest fusion similarity, and determine the second user with the maximum fusion similarity as the first user.

The identification scheme provided by the present disclosure integrates the spatio-temporal information of the user entering or leaving the target area during the process of user identification, so that the identification of the user's identity is more accurate. Compared with single use of face recognition for identity recognition, the present disclosure is more suitable for rail transit scenarios with large-scale data, and can effectively reduce the false alarm of similar users caused by the increase in the order of magnitude of collected images. In addition, as time goes by, due to the continuous accumulation of historical big data, the accuracy of identification provided by the present disclosure can be iteratively enhanced gradually. At the same time, for the false detection that may occur in single face recognition due to the lack of similarity between the images in the database and the on-site pictures, the robustness of identity recognition can also be improved by combining the historical spatiotemporal distribution of historical users.

It can be understood that the above-mentioned method embodiments mentioned in this disclosure can all be combined with each other to form a combined embodiment without violating the principle and logic. Due to space limitations, this disclosure will not repeat them. Those skilled in the art can understand that, in the above method in the specific implementation manner, the specific execution order of each step should be determined according to its function and possible internal logic.

In addition, this disclosure also provides identification devices, electronic equipment, computer-readable storage media, and program products, all of which can be used to implement any of the identification methods provided by this disclosure, and refer to the corresponding technical solutions and descriptions in the method section. record, no more details.

Fig. 5 shows a block diagram of an identity recognition device according to an embodiment of the present disclosure. As shown in Fig. 5, the device includes:

An acquisition module 31, configured to acquire the collected images and spatio-temporal information of the first user, wherein the spatio-temporal information is used to indicate the collection time and collection location of the collected images;

The first determining module 32 is configured to determine at least one second user in the user database according to the captured image of the first user;

The second determining module 33 is configured to respectively determine the historical spatio-temporal distribution of the at least one second user, wherein the historical spatio-temporal distribution is used to indicate that the second user appears in one or more places in at least one time period probability;

An identifying module 34, configured to identify the first user among the at least one second user based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user.

In one or more possible implementation manners, the identification module 34 is configured to determine, based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, the The spatiotemporal similarity of the at least one second user; based on the spatiotemporal similarity between the first user and the at least one second user, identifying the first user among the at least one second user; or , identifying the first user among the at least one second user based on the visual similarity between the first user and the at least one second user and the spatio-temporal similarity respectively.

In one or more possible implementations, the identification module 34 is configured to, for at least one second user, determine that the second user is in the collected The probability that time is located at the collection location; determine the spatio-temporal similarity between the first user and the second user according to the probability.

In one or more possible implementation manners, the identification module 34 is configured to determine the second user in the outbound distribution of the second user when the collection location is an outbound site The probability that the time period corresponding to the collection time is located at the outbound site; or, in the case that the collection site is an inbound site, in the inbound distribution of the second user, determine the second The probability that the user is located at the inbound site during the time period corresponding to the collection time.

In one or more possible implementations, the recognition module 34 is configured to, for at least one second user, weight the visual similarity and spatiotemporal similarity between the first user and the second user , to obtain the fusion similarity between the first user and the second user; based on the fusion similarity between the first user and the at least one second user, identify the first user.

In one or more possible implementations, the identification module 34 is configured to determine, among the at least one second user, the second user whose fusion similarity with the first user is the largest as the first user The second user that the user matches.

In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the methods described in the method embodiments above, and its specific implementation can refer to the description of the method embodiments above. For brevity, here No longer.

Embodiments of the present disclosure also provide a computer-readable storage medium, on which computer program instructions are stored, and the above-mentioned method is implemented when the computer program instructions are executed by a processor. Computer readable storage media may be volatile or nonvolatile computer readable storage media.

An embodiment of the present disclosure also proposes an electronic device, including: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to invoke the instructions stored in the memory to execute the above method.

An embodiment of the present disclosure also provides a computer program product, including computer-readable codes, or a non-volatile computer-readable storage medium carrying computer-readable codes, when the computer-readable codes are stored in a processor of an electronic device When running in the electronic device, the processor in the electronic device executes the above method.

Electronic devices may be provided as terminals, servers, or other forms of devices.

FIG. 6 shows a block diagram of an electronic device 800 according to an embodiment of the present disclosure. For example, the electronic device 800 may be a terminal such as a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, or a personal digital assistant.

6, electronic device 800 may include one or more of the following components: processing component 802, memory 804, power supply component 806, multimedia component 808, audio component 810, input/output (I/O) interface 812, sensor component 814 , and the communication component 816.

The processing component 802 generally controls the overall operations of the electronic device 800, such as those associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above method. Additionally, processing component 802 may include one or more modules that facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802 .

The memory 804 is configured to store various types of data to support operations at the electronic device 800 . Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phonebook data, messages, pictures, videos, and the like. The memory 804 can be implemented by any type of volatile or non-volatile storage device or their combination, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic or Optical Disk.

The power supply component 806 provides power to various components of the electronic device 800 . Power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for electronic device 800 .

The multimedia component 808 includes a screen providing an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may not only sense a boundary of a touch or swipe action, but also detect a duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 808 includes a front camera and/or a rear camera. When the electronic device 800 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a microphone (MIC), which is configured to receive external audio signals when the electronic device 800 is in operation modes, such as call mode, recording mode and voice recognition mode. Received audio signals may be further stored in memory 804 or sent via communication component 816 . In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and a peripheral interface module, which may be a keyboard, a click wheel, a button, and the like. These buttons may include, but are not limited to: a home button, volume buttons, start button, and lock button.

Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of electronic device 800 . For example, the sensor component 814 can detect the open/closed state of the electronic device 800, the relative positioning of components, such as the display and the keypad of the electronic device 800, the sensor component 814 can also detect the electronic device 800 or a Changes in position of components, presence or absence of user contact with electronic device 800 , electronic device 800 orientation or acceleration/deceleration and temperature changes in electronic device 800 . Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. Sensor assembly 814 may also include an optical sensor, such as a complementary metal-oxide-semiconductor (CMOS) or charge-coupled device (CCD) image sensor, for use in imaging applications. In some embodiments, the sensor component 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor or a temperature sensor.

The communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices. The electronic device 800 can access a wireless network based on a communication standard, such as a wireless network (WiFi), a second generation mobile communication technology (2G) or a third generation mobile communication technology (3G), or a combination thereof. In an exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology and other technologies.

In an exemplary embodiment, electronic device 800 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A programmable gate array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation for performing the methods described above.

In an exemplary embodiment, there is also provided a non-volatile computer-readable storage medium, such as the memory 804 including computer program instructions, which can be executed by the processor 820 of the electronic device 800 to implement the above method.

FIG. 7 shows a block diagram of an electronic device 1900 according to an embodiment of the present disclosure. For example, electronic device 1900 may be provided as a server. Referring to FIG. 7 , electronic device 1900 includes processing component 1922 , which further includes one or more processors, and a memory resource represented by memory 1932 for storing instructions executable by processing component 1922 , such as application programs. The application programs stored in memory 1932 may include one or more modules each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input-output (I/O) interface 1958 . The electronic device 1900 can operate based on the operating system stored in the memory 1932, such as the Microsoft server operating system (Windows Server ^TM ), the graphical user interface-based operating system (Mac OS X ^TM ) introduced by Apple Inc., and the multi-user and multi-process computer operating system (Unix ^™ ), a free and open source Unix-like operating system (Linux ^™ ), an open source Unix-like operating system (FreeBSD ^™ ), or the like.

In an exemplary embodiment, there is also provided a non-transitory computer-readable storage medium, such as the memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to implement the above method.

The present disclosure can be a system, method and/or computer program product. A computer program product may include a computer readable storage medium having computer readable program instructions thereon for causing a processor to implement various aspects of the present disclosure.

A computer readable storage medium may be a tangible device that can retain and store instructions for use by an instruction execution device. A computer readable storage medium may be, for example, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory), static random access memory (SRAM), compact disc read only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanically encoded device, such as a printer with instructions stored thereon A hole card or a raised structure in a groove, and any suitable combination of the above. As used herein, computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., pulses of light through fiber optic cables), or transmitted electrical signals.

Computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or downloaded to an external computer or external storage device over a network, such as the Internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers. A network adapter card or a network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .

Computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or Source or object code written in any combination, including object-oriented programming languages - such as Smalltalk, C++, etc., and conventional procedural programming languages - such as the "C" language or similar programming languages. Computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as via the Internet using an Internet service provider). connect). In some embodiments, an electronic circuit, such as a programmable logic circuit, field programmable gate array (FPGA), or programmable logic array (PLA), can be customized by utilizing state information of computer-readable program instructions, which can Various aspects of the present disclosure are implemented by executing computer readable program instructions.

Aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It should be understood that each block of the flowcharts and/or block diagrams, and combinations of blocks in the flowcharts and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine such that when executed by the processor of the computer or other programmable data processing apparatus , producing an apparatus for realizing the functions/actions specified in one or more blocks in the flowchart and/or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause computers, programmable data processing devices and/or other devices to work in a specific way, so that the computer-readable medium storing instructions includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks in flowcharts and/or block diagrams.

It is also possible to load computer-readable program instructions into a computer, other programmable data processing device, or other equipment, so that a series of operational steps are performed on the computer, other programmable data processing device, or other equipment to produce a computer-implemented process , so that instructions executed on computers, other programmable data processing devices, or other devices implement the functions/actions specified in one or more blocks in the flowcharts and/or block diagrams.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, a portion of a program segment, or an instruction that includes one or more Executable instructions. In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It should also be noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by a dedicated hardware-based system that performs the specified function or action , or may be implemented by a combination of dedicated hardware and computer instructions.

The computer program product can be specifically realized by means of hardware, software or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK) etc. Wait.

Having described various embodiments of the present disclosure above, the foregoing description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and alterations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principle of each embodiment, practical application or improvement of technology in the market, or to enable other ordinary skilled in the art to understand each embodiment disclosed herein.

Claims

An identification method, characterized in that it comprises:

Acquiring the collected images and spatio-temporal information of the first user, wherein the spatio-temporal information is used to indicate the collection time and collection location of the collected images;

determining at least one second user in a user database according to the captured image of the first user;

respectively determining the historical spatio-temporal distribution of the at least one second user, wherein the historical spatio-temporal distribution is used to represent the probability that the second user appears in one or more places during at least one time period;

Identifying the first user among the at least one second users based on the spatiotemporal information of the first user and the historical spatiotemporal distribution of the at least one second user.
The method according to claim 1, characterized in that, based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, identifying the second user among the at least one second user a user, including:

Based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, determine the spatio-temporal similarity between the first user and the at least one second user;

identifying said first user among said at least one second user based on said spatiotemporal similarity between said first user and said at least one second user, respectively; or

Identifying the first user among the at least one second users based on the visual similarity and the spatio-temporal similarity between the first user and the at least one second user respectively.
The method according to claim 2, wherein, based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user, it is determined that the first user is related to the at least one second user respectively. The spatio-temporal similarity of two users, including:

For at least one second user, in the historical spatio-temporal distribution of the second user, determine the probability that the second user is located at the collection location at the collection time;

determining the spatio-temporal similarity between the first user and the second user according to the probability.
The method according to claim 3, wherein the historical spatiotemporal distribution includes an inbound distribution and an outbound distribution, and in the historical spatiotemporal distribution of the second user, it is determined that the second user is in the Probability that the collection time is located at the collection location, including:

In the case where the collection location is an outbound site, in the outbound distribution of the second user, determine the probability that the second user is located at the outbound site during the time period corresponding to the collection time; or ,

In a case where the collection site is an inbound site, in the inbound distribution of the second user, determine the probability that the second user is located in the inbound site in a time period corresponding to the collection time.
The method according to any one of claims 2 to 4, wherein, based on the visual similarity and the spatio-temporal similarity between the first user and the at least one second user respectively, in the at least identifying said first user among a second user, comprising:

For at least one of the second users, weighting the visual similarity and spatio-temporal similarity between the first user and the second user to obtain the fused similarity between the first user and the second user;

Identifying the first user among the at least one second user based on fusion similarities between the first user and the at least one second user respectively.
The method according to claim 5, wherein the first user is identified among the at least one second user based on the fused similarities between the first user and the at least one second user respectively ,include:

Determining, among the at least one second user, a second user having the greatest fusion similarity with the first user as the second user matched with the first user.
The method according to any one of claims 1 to 6, wherein the determining at least one second user in the user database according to the captured image of the first user includes:

Matching the captured image of the first user with the user images of historical users in the user database to determine the visual similarity between the first user and the historical user;

If the visual similarity is greater than a preset threshold, the historical user is determined as the second user.
The method according to any one of claims 1 to 7, wherein the method further comprises:

acquiring historical user records of a plurality of historical users, wherein the plurality of historical users includes the at least one second user;

Based on the historical user records, count the historical times that the historical users appear in one or more locations during at least one time period, and count the historical times that one or more historical users appear in the one or more locations during the at least one time period total number of locations;

According to the historical times and the total times, the historical spatio-temporal distribution of the historical users is generated.
The method according to claim 8, characterized in that the method further comprises:

Smoothing is performed on the historical spatio-temporal distribution to obtain the historical spatio-temporal distribution of the historical users after smoothing processing.
The method according to any one of claims 1 to 9, wherein after identifying the first user among the at least one second user, further comprising:

saving the captured image and spatio-temporal information of the first user in the historical user record of the first user, wherein the historical user record includes at least one of the following information: user identification, inbound identification or outbound identification, Station ID, time of inbound or outbound.
An identification device, characterized in that it comprises:

An acquisition module, configured to acquire the collected image and spatio-temporal information of the first user, wherein the spatio-temporal information is used to indicate the collection time and location of the collected image;

A first determining module, configured to determine at least one second user in the user database according to the captured image of the first user;

A second determining module, configured to respectively determine the historical spatiotemporal distribution of the at least one second user, wherein the historical spatiotemporal distribution is used to represent the probability that the second user appears in one or more places in at least one time period ;

An identifying module, configured to identify the first user among the at least one second user based on the spatio-temporal information of the first user and the historical spatio-temporal distribution of the at least one second user.
An electronic device, characterized in that it comprises:

processor;

memory for storing processor-executable instructions;

Wherein, the processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1-10.
A computer-readable storage medium, on which computer program instructions are stored, wherein, when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented.
A computer program product, comprising computer readable codes, or a non-volatile computer readable storage medium bearing computer readable codes, when the computer readable codes are run in a processor of an electronic device, the electronic A processor in the device is configured to implement the method of any one of claims 1-10.