Data processing method, device, computer equipment and storage medium

Abstract

An embodiment of the present application discloses a data processing method, apparatus, computer device, and storage medium. The present application belongs to the computer vision technology in the field of artificial intelligence. The data processing method includes: obtaining a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period; obtaining a video stream including the user to be verified, where the user to be verified refers to the user waiting for identity verification who enters the detection area during the detection time period; determining the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream; and determining the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tag device. By using the present application, the efficiency and accuracy of identity verification can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular, to a data processing method, apparatus, computer device, and storage medium. Background Art

[0002] With the highly developed information technology, computer-based network data centers are more and more widely used in various fields. The computer room is the infrastructure of the network data platform center. Computer devices, server devices, network devices, storage devices, etc. are the core devices of the data center computer room. Based on these core devices, centralized processing, storage, transmission, exchange, management, etc. of information can be realized in a physical space.

[0003] In order to protect the security of the data center computer room, currently, manual identity verification is performed on the access personnel entering and leaving the data center computer room, and only legal access personnel who pass the manual verification are allowed to enter the data center computer room. It can be seen that there is a problem with the verification efficiency of manual verification one by one, resulting in low verification efficiency of identity verification; and because manual verification is subject to subjective influence, it will also lead to low accuracy of identity verification. Summary of the Invention

[0004] Embodiments of this application provide a data processing method, apparatus, computer device, and storage medium, which can improve the efficiency and accuracy of identity verification.

[0005] Embodiments of this application on the one hand provide a data processing method, including:

[0006] Obtain a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period;

[0007] Obtain a video stream including the user to be verified, where the user to be verified refers to the user waiting for identity verification who enters the detection area during the detection time period;

[0008] Determine the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream;

[0009] Determine the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tag device.

[0010] Embodiments of this application on the one hand provide a data processing apparatus, including:

[0011] A first acquisition module, configured to acquire a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period;

[0012] A second acquisition module, configured to acquire a video stream including a user to be verified, where the user to be verified refers to a user waiting for identity verification who enters a detection area during a detection time period;

[0013] An identification module, configured to determine the relationship between the user to be verified and a user label device according to a reference user identity and the video stream;

[0014] A first determination module, configured to determine an identity verification result of the user to be verified according to the relationship between the user to be verified and the user label device.

[0015] On the one hand, an embodiment of the present application provides a computer device, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor executes the methods in the foregoing embodiments.

[0016] On the one hand, an embodiment of the present application provides a computer storage medium. The computer storage medium stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the methods in the foregoing embodiments are executed.

[0017] On the one hand, an embodiment of the present application provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions. The computer instructions are stored in a computer-readable storage medium. When the computer instructions are executed by a processor of a computer device, the methods in the foregoing embodiments are executed.

[0018] The present application determines an identity verification result of a user to be verified according to a reference user identity corresponding to a user label device located in a specified detection area during a specified detection time period, and a video stream of the user to be verified who enters the specified detection area during the specified detection time period. It can be seen that the present application automatically performs identity verification by a terminal device without manual participation, which can improve the verification efficiency of identity verification; and the terminal device automatically identifies without subjective factor interference, which can ensure the accuracy of identity verification; further, by determining the relationship between the user and the user label device by judging the user label device and the video stream located in the same area during the same time period, for the situation where the user label device and the user are separated, the solution of the present application can also accurately identify it. Description of the Drawings

[0019] Figure 1 is a system architecture diagram of data processing provided by an embodiment of the present application;

[0020] Figures 2a - 2d is a schematic diagram of a data processing scenario provided by an embodiment of the present application;

[0021] Figure 3 is a flowchart of a data processing method provided by an embodiment of the present application;

[0022] Figure 4 is a schematic flowchart of a data processing method provided by an embodiment of the present application;

[0023] Figure 5a is a schematic flowchart of a process for determining the identity of a reference user provided by an embodiment of the present application;

[0024] Figure 5b is a schematic diagram for determining the positioning coordinates of a user tag device provided by an embodiment of the present application;

[0025] Figure 6 is an interaction diagram for detecting the separation of a person and a card provided by an embodiment of the present application;

[0026] Figure 7 is an interaction diagram for detecting the separation of a person and a card provided by an embodiment of the present application;

[0027] Figure 8 is an interaction diagram for detecting the separation of a person and a card provided by an embodiment of the present application;

[0028] Figure 9 is an interaction diagram for detecting the separation of a person and a card provided by an embodiment of the present application;

[0029] Figure 10 is a schematic structural diagram of a data processing device provided by an embodiment of the present application;

[0030] Figure 11 is a schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0031] Next, the technical solutions in the embodiments of the present application will be described with reference to the accompanying drawings in the embodiments of the present application.

[0032] Artificial Intelligence (AI) uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, a theory, method, technology, and application system that perceives the environment, acquires knowledge, and uses knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a way similar to human intelligence. Artificial intelligence also studies the design principles and implementation methods of various intelligent machines to enable the machines to have the functions of perception, reasoning, and decision-making.

[0033] Artificial intelligence technology is an interdisciplinary subject with a wide range of fields, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, large image processing technology, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0034] This application relates to computer vision technology (CV) under artificial intelligence, specifically belonging to video content recognition in computer vision technology, and specifically relates to identifying the user identity of users in a video stream.

[0035] Computer vision technology is a science that studies how to make machines "see". More specifically, it refers to using cameras and computers to replace the human eye to perform machine vision such as target recognition, tracking, and measurement on targets, and further performing graphic processing to make the computer process into an image that is more suitable for human eye observation or transmitted to an instrument for detection. As a scientific discipline, computer vision studies related theories and technologies, and attempts to establish an artificial intelligence system that can obtain information from images or multi-dimensional data. Computer vision technology usually includes technologies such as image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, and video content / behavior recognition.

[0036] This application can be applied to the following scenarios: when it is necessary to identify whether the access personnel entering the computer room wear their own work cards, obtain the reference user identity corresponding to the work card in the specified scene area within the specified time period, obtain the video stream of the user to be verified entering the specified scene area from the camera within the specified time period, determine the user identity of the user to be verified by identifying the video stream, and determine whether the user to be verified wears their own work card according to the user identity of the user to be verified and the reference user identity, and further determine whether the user to be verified passes the identity verification.

[0037] Subsequently, if it is determined that the user to be verified does not wear their own work card (that is, the user to be verified fails the identity verification), an alarm message can be sent to remind the user to be verified and avoid the occurrence of computer room security incidents and protect the computer room security.

[0038] Please refer to Figure 1 , Figure 1 which is a system architecture diagram of data processing provided by an embodiment of this application. This application relates to a terminal device 10a, a camera 10b, a positioning base station 10c, and a server 10d, etc., and the camera 10b and the positioning base station 10c can be installed in the computer room.

[0039] The positioning base station 10c can obtain the positioning coordinates of the work card devices that appear in the computer room at different time periods, and send the obtained positioning coordinates and the time period corresponding to each positioning coordinate to the server 10d. The camera 10b can collect the video streams of the users entering different areas of the computer room at different time periods, and send the collected video streams to the server 10d.

[0040] The server 10d can screen out the reference user identities corresponding to the work card devices that appear in the specified area of the computer room within the specified time period, and determine the to-be-verified user identities of the to-be-verified users who enter the specified area of the computer room within the specified time period according to the video streams. The server 10d determines whether the to-be-verified users wear their own work card devices by judging whether the reference user identities are consistent with the to-be-verified user identities, and further determines the identity verification results of the to-be-verified users.

[0041] The terminal device 10a, the camera 10b, and the positioning base station 10c can be directly or indirectly connected to the server 10d through wired or wireless communication methods, and this application does not make any restrictions here.

[0042] The following takes how the server 10d identifies whether the users entering the computer room wear their own work cards as an example for detailed description. Please refer to Figures 2a - 2d which is a schematic diagram of a data processing scenario provided by an embodiment of this application. Figure 2a In the image 20a, it is a floor plan of the computer room. It can be seen from the floor plan 20a of the computer room that the entire computer room is divided into 4 areas, namely Area 1, Area 2, Area 3, and Area 4, and cameras are installed in each area. The cameras in each area can capture the video streams in the corresponding areas.

[0043] In addition to installing cameras in the computer room, UWB (Ultra Wide Band) positioning base stations are also installed. The employees entering the computer room will wear work cards, and the chips in the work cards will regularly send positioning signals and work card numbers to the UWB positioning base stations. The UWB positioning base stations can determine the positioning coordinates of the work cards that appear in the computer room at different times according to the received positioning signals. The UWB positioning base stations can send the determined positioning coordinates of the work cards that appear in the computer room at different times and the corresponding work card numbers to the server 10d.

[0044] From Figure 2a the positioning table 20b in it, it can be known that the positioning coordinates of the work card 1 corresponding to the work card number 1 detected by the UWB positioning base station at the two times of 08:10 and 08:11 are both (x1, y1).

[0045] From Figure 2aAs can be seen from the positioning table 20c in , the UWB positioning base station detected that the positioning coordinates of the work card 2 corresponding to the work card number 2 at 08:10 and 08:11 are both (x1, y1).

[0046] The UWB positioning base station can send the above-obtained data to the server 10d. By analyzing the positioning coordinates "(x1, y1)", the server 10d can determine that the positioning coordinates "(x1, y1)" are located in Area 2 of the computer room.

[0047] The server 10d can generate an identity detection request for a region at regular intervals. The identity detection request includes the detection time period and the detection region. Assuming that the detection time period in the current identity detection request is [08:10 - 08:11] and the detection region is Area 2, then the server 10d can determine, based on the data reported by the UWB positioning base station, that the work cards in Area 2 during the time period [08:10 - 08:11] are Work Card 1 and Work Card 2.

[0048] Since the cameras in the computer room can continuously collect video streams in different areas of the computer room, the server 10d can also obtain the video stream in Area 2 during the time period [08:10 - 08:11] through the cameras. As Figure 2b shown, assuming that the server 10d obtains the video stream 20d, the server 10d can identify the users in the video stream 20d by calling the trained face recognition model. Assuming that the face recognition model recognizes that the video stream 20d includes Employee 1 and Employee 2.

[0049] The server 10d can query whether the work card of Employee 1 is Work Card 1 and whether the work card of Employee 2 is Work Card 2. Assuming that the server 10d queries that the work card of Employee 1 is Work Card 1 and the work card of Employee 2 is Work Card 2, then the server 10d can determine that both Employee 1 and Employee 2 wore their own work cards during the time period 08:10 - 08:11, and it can be considered that the identity verification of Employee 1 and Employee 2 has passed at this time; conversely, if the server 10d queries that Employee 1 does not correspond to Work Card 1 (or Employee 2 does not correspond to Work Card 2), then the server 10d can determine that Employee 1 (or Employee 2) did not wear their own work card during the time period 08:10 - 08:11, and it can be considered that the identity verification of Employee 1 (or Employee 2) has not passed at this time.

[0050] Employees can move around freely in the computer room, and the UWB positioning base station can continuously detect the positioning coordinates of the work card at different times. As can be seen from Figure 2a the positioning table 20b in , the UWB positioning base station detected that the positioning coordinates of Work Card 1 at 08:12 and 08:13 are both (x2, y2).

[0051] As can be seen from Figure 2aAs can be seen from the positioning table 20c, the UWB positioning base station detected that the positioning coordinates of the work card 2 at 08:12 and 08:13 are (x3, y3).

[0052] The UWB positioning base station can send the above-obtained data to the server 10d. By analyzing the positioning coordinates "(x2, y2)", the server 10d can determine that the positioning coordinates "(x2, y2)" are located in area four; by analyzing the positioning coordinates "(x3, y3)", the server 10d can determine that the positioning coordinates "(x3, y3)" are located in area one.

[0053] The server 10d can also generate an identity detection request for a region at regular intervals. Assuming that the detection time period in the current identity detection request is [08:12 - 08:13] and the detection region is area one, then the server 10d can determine the work card 2 that is located in area one during the time period [08:12 - 08:13] based on the data reported by the UWB positioning base station.

[0054] The server 10d can also obtain the video stream located in area one during the time period [08:12 - 08:13] through the camera. As Figure 2c shown, assuming that the server 10d obtains the video stream 20e, the server 10d calls the trained face recognition model to recognize the users in the video stream 20e. Assuming that the face recognition model recognizes that the video stream 20d includes employee 1.

[0055] The server 10d can query whether the work card of employee 1 is work card 2. If the server 10d queries that the work card of employee 1 is work card 1, then the server 10d can determine that employee 1 did not wear his own work card during the time period 08:12 - 08:13. At this time, the identity verification of employee 1 fails. The server 10d can output an alarm message to prompt employee 1 to wear his own work card.

[0056] Similarly, as Figure 2d shown, the server 10d can also generate an identity detection request for area four. Assuming that the detection time period in the current identity detection request is [08:12 - 08:13] and the detection region is area four, then the server 10d can determine the work card 1 that is located in area four during the time period [08:12 - 08:13] based on the data reported by the UWB positioning base station.

[0057] The server 10d can also obtain the video stream located in area four during the time period [08:12 - 08:13] through the camera. As Figure 2cAs shown in the figure, assume that the server 10d obtains the video stream 20f. The server 10d identifies the users in the video stream 20f by calling the trained face recognition model. Assume that the face recognition model identifies that the video stream 20d includes employee 2.

[0058] The server 10d can query whether the work card of employee 2 is work card 1. If the server 10d queries that the work card of employee 2 is work card 2, then the server 10d can determine that employee 2 did not wear his own work card during the time period from 08:12 to 08:13. At this time, the identity verification of employee 2 fails. The server 10d can output an alarm message to prompt employee 2 to wear his own work card.

[0059] As can be seen from the above, in area two, both employee 1 and employee 2 wore their own work cards. However, later, employee 1 and employee 2 exchanged work cards, which led to employee 1 being detected as not wearing his own work card when entering area one, and employee 2 being detected as not wearing his own work card when entering area four.

[0060] Among them, the specific process of obtaining the reference user identity and the video stream (such as the video stream 20d, or video stream 20e or video stream 20f in the above embodiments) can be referred to in the following Figures 3 - 9 corresponding embodiments.

[0061] Please refer to Figure 3 , Figure 3 is a schematic flowchart of a data processing method provided by an embodiment of the present application. The following embodiments are described with a server with better performance (such as the server 10d in the above Figures 2a - 2c corresponding embodiments) as the execution subject. The data processing method includes the following steps:

[0062] Step S101, obtain the reference user identity. The reference user identity refers to the user identity corresponding to the user label device located in the detection area during the detection time period.

[0063] Specifically, the server obtains the label device reporting request (such as the identity detection request in the above Figures 2a - 2d corresponding embodiments). The label device reporting request includes the detection time period and the detection area. The server obtains the device identifier of the user label device located in the detection area during the detection time period.

[0064] The user tag device can be a work card. One user tag device corresponds to one user, and the user tag device can send a positioning signal, which can be used to determine the tag positioning information of the user tag device. The server determines the user identity corresponding to the device identifier of the obtained user tag device (referred to as the reference user identity) by querying the database. Among them, the user identity can be a user ID, or a user ID + user name. Generally speaking, the user identity can uniquely identify a user, that is, the user identity has uniqueness and exclusivity.

[0065] The tag device reporting request can be generated by the server according to a preset rule. For example, a tag device reporting request is generated every 10 minutes. The tag device reporting request can also be determined by the server according to the tag positioning information of the received user tag device. For example, the tag device currently receives the tag positioning information of the user tag device in the time interval from 08:00 to 09:00, and the positioning coordinates in the tag positioning information are located in Detection Area 1. Therefore, the server can generate a tag device reporting request in real time. At this time, the detection time period in the tag device reporting request is 08:00 - 09:00, and the detection area is Detection Area 1. The tag device reporting request can also be determined by the server according to the received video stream. For example, the tag device currently receives the video stream in Detection Area 1, and the start timestamp of the video stream is 08:00 and the end timestamp is 09:00. Therefore, the server can generate a tag device reporting request in real time. At this time, the detection time period in the tag device reporting request is 08:00 - 09:00, and the detection area is Detection Area 1.

[0066] Step S102: Obtain a video stream containing the user to be verified. The user to be verified refers to a user who enters the detection area during the detection time period and waits for identity verification.

[0067] Specifically, the server obtains the video stream collected during the detection time period and in the detection area (such as video stream 20d, or video stream 20e, or video stream 20f in the corresponding embodiment above). The video stream includes the user to be verified. Of course, the user to be verified is the user who enters the detection area during the detection time period and waits for verification. Figures 2a - 2d

[0068] Step S103: Determine the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream.

[0069] Specifically, the server performs video recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified. Among them, the video recognition processing can include face recognition processing and / or human body recognition processing. Face recognition processing refers to identifying the user identity of the user to be verified, and human body recognition processing refers to identifying the number of users of the user to be verified.

[0070] The specific processes of face recognition processing and human body recognition processing will be described separately below.

[0071] First, the specific process of face recognition processing will be described:

[0072] The video stream includes multiple video frame images. In each video frame image, a face region is determined. Among them, the pixel type of each pixel in each video frame image can be recognized through a semantic segmentation model. The pixel type includes face type or background type. In each video frame image, the smallest connected region composed of pixels whose pixel type belongs to the face type is used as the face region. The face regions of all video frame images can be called the face region of the video stream. If there is only one face region in each video frame image, the server extracts the face region of each video frame image to obtain a pure face image. The face features of each face image are extracted through a feature extraction model. Among them, the feature extraction model can be a unit model in an expression classification model. The expression recognition model can include, in addition to the feature extraction model, an expression type recognition model. That is, the feature extraction model is used to extract facial features, and the expression type recognition model is used to recognize the expression type corresponding to the feature based on the extracted facial features.

[0073] The server can superimpose the face features of all face images to obtain the face features to be verified of the video stream.

[0074] If there are multiple face regions in a video frame image, the server needs to align multiple video frame images to determine the face region of each face in each video frame image. The face images and the face features of the face images are extracted in the same way. The server superimposes the human features belonging to the same face respectively, and multiple face features to be verified of the video stream can be obtained.

[0075] Optionally, in addition to determining the face features to be verified of the video stream in the above way, the server can also determine the face features to be verified of the video stream in the following way: The video stream includes multiple video frame images. The server extracts one video frame image (referred to as the target video frame image) from these multiple video frame images. The face region in the target video frame image is recognized through a semantic segmentation model, and the face region of the target video frame image is the face region of the video stream. The server extracts the face region of the target video frame image to obtain a face image, and extracts the face features of the face image (referred to as the face features to be verified) by calling the feature extraction model. And the face features to be verified extracted from the target video frame image are the face features to be verified of the video stream.

[0076] After the server determines the face features to be verified in the video stream, it matches the face features to be verified with the face features of multiple user identities in the user identity database. In the user identity database, the face features that match the face features to be verified are used as the face features to be extracted. The user identity corresponding to the face features to be extracted is obtained in the user database, and the extracted user identity is used as the user identity to be verified. The server uses the user identity to be verified as the information of the user to be verified for the user to be verified. The face features that match the face features to be verified refer to the case where the feature distance between the face features to be verified and the face features is less than the distance threshold.

[0077] Next, the specific process of human body recognition processing is described:

[0078] The video stream includes multiple video frame images. The server performs semantic segmentation processing on each video frame image by calling a semantic segmentation model to obtain the pixel type of each pixel in each video frame image. The pixel type includes a human body type or a background type. The server uses the smallest connected region composed of pixels with a human body type as the human body region. Thus, the human body region in each video frame image can be determined. The number of regions of the human body region in each video frame image is counted, and the average value of the number of regions of all video frame images is used as the number of users of the user to be verified. The server uses this number of users as the information of the user to be verified for the user to be verified.

[0079] Optionally, the server can perform both face recognition processing and human body recognition processing on the video stream, so as to obtain the user identity to be verified of the user to be verified and the number of users of the user to be verified.

[0080] It should be noted that the reason for using face recognition processing or human body recognition processing to determine the information of the user to be verified in this application is that under certain conditions, it may only be possible to perform human body recognition processing and not face recognition processing. For example, when the light of the video stream is relatively dim, if face recognition processing is used, the user identity to be verified cannot be recognized, and only human body recognition can be performed, that is, the number of users can be recognized. Of course, when the light of the video stream is relatively bright, according to business requirements, face recognition processing or human body recognition processing can be performed, or both face recognition processing and human body recognition processing can be performed simultaneously.

[0081] After the server determines the information of the user to be verified for the user to be verified, if the information of the user to be verified includes the user identity to be verified, the server determines whether the user identity to be verified is the same as the reference user identity. If they are the same, it can be determined that the relationship between the user to be verified and the user label device is a matching relationship; otherwise, if they are different, it can be determined that the relationship between the user to be verified and the user label device is a non-matching relationship.

[0082] If the user information to be verified includes the number of users to be verified, the server determines whether the number of users is the same as the number of reference user identities. If they are different, it can be determined that the relationship between the user to be verified and the user tag device is a mismatched relationship.

[0083] If the relationship between the user to be verified and the user tag device is a matching relationship, it means that the user to be verified is carrying their own user tag device during the detection time period and in the detection area. Conversely, if the relationship between the user to be verified and the user tag device is a mismatched relationship, it means that the user to be verified is not carrying their own user tag device during the detection time period and in the detection area. Generally speaking, the situation of people-card separation can be detected.

[0084] Step S104, determine the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tag device.

[0085] Specifically, if the relationship between the user to be verified and the user tag device is a matching relationship, it is determined that the identity verification result of the user to be verified is verification passed; if the relationship between the user to be verified and the user tag device is a mismatched relationship, it is determined that the identity verification result of the user to be verified is verification failed.

[0086] Optionally, if the identity verification result of the user to be verified is verification failed, the server can output an alarm message for the user to be verified. Among them, the alarm can be played in the detection area through a speaker, or the alarm message can be sent to the user terminal of the user to be verified (for example, a user's mobile phone or a personal computer).

[0087] As can be seen from the above, based on different lighting conditions, the present application performs human body detection processing or face recognition processing on the video stream, which can ensure the accuracy of the video recognition result, thereby reducing the misjudgment of identity verification; furthermore, when it is detected that the user to be verified is an illegal user, an alarm message is output. Subsequently, corresponding protection measures can be taken based on the alarm message to reduce the risk of economic losses.

[0088] Please refer to Figure 4 , Figure 4 which is a schematic flowchart of a data processing method provided by an embodiment of the present application. This embodiment mainly describes how to determine the relationship between the user to be verified and the user tag device. The data processing method includes the following steps:

[0089] Step S201, obtain the reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period.

[0090] Step S202, obtain the video stream including the user to be verified, where the user to be verified refers to the user waiting for identity verification who enters the detection area during the detection time period.

[0091] Among them, for the specific processes of steps S201 - S202, reference can be made to steps S101 - S102 in the corresponding embodiment above. Figure 3 Corresponding to steps S101 - S102 in the embodiment.

[0092] Step S203: Perform video recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified.

[0093] Specifically, as can be seen from the foregoing, if face recognition processing is performed on the video stream, the information of the user to be verified of the user to be verified may include the identity of the user to be verified of the user to be verified; if human body recognition processing is performed on the video stream, the information of the user to be verified of the user to be verified may include the number of users of the user to be verified.

[0094] Step S204: Determine the relationship between the user to be verified and the user tagging device according to the information of the user to be verified and the reference user identity.

[0095] Specifically, if the information of the user to be verified is generated after performing face recognition processing on the video stream, then the verification process for the user to be verified is as follows:

[0096] Obtain the number of identities of the user to be verified. Assume that the number of identities of the user to be verified is M, and the number of reference user identities is N. If the number of reference user identities N is not equal to the number of identities of the user to be verified M, it is determined that the reference user identity is different from the identity of the user to be verified, and further, the relationship between the user to be verified and the user tagging device can be determined as a mismatch relationship.

[0097] The scenario of this situation is as follows: within the detection time period and in the detection area, 4 work cards are detected, but from the video stream, it is detected that 5 users to be verified enter the detection area within the detection time period. It can be judged that there is one user to be verified not wearing a work card, so it is directly determined that these 5 users to be verified do not match the 4 work cards.

[0098] If the number of reference user identities N is equal to the number of identities of the user to be verified M, the server further detects whether there is a one - to - one correspondence between the N reference user identities and the M identities of the user to be verified. The one - to - one correspondence means that each reference user identity among the N reference user identities exactly pairs with a certain identity of the user to be verified among the M identities of the user to be verified, and each identity of the user to be verified among the M identities of the user to be verified also exactly pairs with a certain reference user identity among the N reference user identities. If one identity of the user to be verified is the same as one reference user identity, it means that this identity of the user to be verified and this reference user identity are paired.

[0099] If the server detects a one-to-one correspondence between N reference user identities and M user identities to be verified, it indicates that the reference user identities and the user identities to be verified are the same. Furthermore, the server can determine that the relationship between the user identities to be verified and the user label device is a matching relationship.

[0100] For example, the user information to be verified includes 3 user identities to be verified, namely the user identity to be verified 1, the user identity to be verified 2, and the user identity to be verified 3. The identity identifier of the user identity to be verified 1 is 001, the identity identifier of the user identity to be verified 2 is 002, and the identity identifier of the user identity to be verified 3 is 003. The number of reference user identities is also 3, namely the reference user identity 1, the reference user identity 2, and the reference user identity 3. The identity identifier of the reference user identity 1 is 001, the identity identifier of the reference user identity 2 is 002, and the identity identifier of the reference user identity 3 is 003. At this time, it shows that there is a one-to-one correspondence between the 3 user identities to be verified and the 3 reference user identities. Therefore, the server can determine that these 3 users to be verified can match these 3 reference user identities.

[0101] The scenario of this situation is as follows: within the detection time period and in the detection area, 4 work cards are detected, and it is detected from the video stream that there are 4 users to be verified entering the detection area within the detection time period, and the employee numbers corresponding to these 4 work cards are the same as the employee numbers of these 4 users to be verified. It can be judged that these 4 users to be verified are all wearing their own work cards. Therefore, it can be determined that these 4 users to be verified and the 4 work cards match.

[0102] If the server detects that there is no one-to-one correspondence between N reference user identities and M user identities to be verified, it indicates that the reference user identities and the user identities to be verified are different. Furthermore, the server can determine that the relationship between the user identities to be verified and the user label device is a non-matching relationship.

[0103] For example, the user information to be verified includes 3 user identities to be verified, namely the user identity to be verified 1, the user identity to be verified 2, and the user identity to be verified 3. The identity identifier of the user identity to be verified 1 is 001, the identity identifier of the user identity to be verified 2 is 002, and the identity identifier of the user identity to be verified 3 is 003. The number of reference user identities is also 3, namely the reference user identity 1, the reference user identity 2, and the reference user identity 3. The identity identifier of the reference user identity 1 is 001, the identity identifier of the reference user identity 2 is 004, and the identity identifier of the reference user identity 3 is 005. At this time, it shows that there is no one-to-one correspondence between the 3 user identities to be verified and the 3 reference user identities. Therefore, the server can determine that these 3 users to be verified are illegal users.

[0104] The scenario of this situation is as follows: within the detection time period and in the detection area, 4 work cards are detected, and 4 users to be verified who enter the detection area during the detection time period are detected from the video stream. And there are at least two employee numbers among the corresponding employee numbers of these 4 work cards and the employee numbers of these 4 users to be verified that do not have a corresponding relationship. It can be determined that there is at least one user to be verified among these 4 users to be verified who is not wearing their own work card (that is, the person and the card are separated). Therefore, it is directly determined that these 4 users to be verified do not match the 4 work cards.

[0105] If the information of the user to be verified is generated after performing human body recognition processing on the video stream, then the verification process for the user to be verified is as follows:

[0106] The information of the user to be verified includes the number of users to be verified. The server obtains the number of reference user identities. If the number of reference user identities is not equal to the number of users to be verified, the server can determine that the relationship between the user to be verified and the user label device is a non-matching relationship.

[0107] The scenario of this situation is as follows: within the detection time period and in the detection area, 4 work cards are detected, but 5 users to be verified who enter the detection area during the detection time period are detected from the video stream. It can be determined that there is one user to be verified who is not wearing a work card. Therefore, it is directly determined that these 5 users to be verified do not match the 4 work cards.

[0108] Step S205, determine the identity verification result of the user to be verified according to the relationship between the user to be verified and the user label device.

[0109] Among them, the specific process of step S205 can refer to step S104 in the above Figure 3 corresponding embodiment.

[0110] As can be seen from the above, when this application determines the identity verification result of the user to be verified, it considers various situations of the separation of the person and the card. Therefore, this application has a high accuracy rate in identifying the separation of the person and the card.

[0111] Please refer to Figure 5a , Figure 5a is a schematic flowchart of a process for determining the identity of a reference user provided by an embodiment of this application. Determining the identity of a reference user includes the following steps S301 - step S303, and steps S301 - step S303 are Figure 3 a specific embodiment corresponding to step S101 in the corresponding embodiment:

[0112] Step S301, obtain a label device reporting request. The label device reporting request includes a detection time period and a detection area.

[0113] Specifically, the server obtains a reporting request from a tag device. The reporting request from the tag device includes a detection time period and a detection area.

[0114] Among them, the tag reporting request can be generated by the server according to a preset rule. For example, a tag device reporting request is generated every 10 minutes.

[0115] Optionally, the tag device reporting request can also be determined by the server based on the received tag positioning information of the user tag device. For example, the tag device currently receives the tag positioning information of the user tag device in the time interval of 08:00 - 09:00, and the positioning coordinates in the tag positioning information are located in Detection Area 1. Therefore, the server can generate a tag device reporting request in real time. At this time, the detection time period in the tag device reporting request is 08:00 - 09:00, and the detection area is Detection Area 1.

[0116] Optionally, the tag device reporting request can also be determined by the server based on the received video stream. For example, the tag device currently receives the video stream in Detection Area 1, and the start timestamp of the video stream is 08:00, and the end timestamp is 09:00. Therefore, the server can generate a tag device reporting request in real time. At this time, the detection time period in the tag device reporting request is 08:00 - 09:00, and the detection area is Detection Area 1.

[0117] Step S302: Obtain multiple tag positioning information within the detection time period. Each tag positioning information includes the device identifier of the original user tag device and the positioning coordinates of the original user tag device.

[0118] Specifically, the server obtains multiple positioning signals from multiple signal acquisition devices. Each positioning signal is sent by an original user tag device. Each positioning signal includes the device identifier of the original user tag device, the reception timestamp, and a signal flag bit. The signal flag bit indicates that the positioning signal is a signal sent by the original user tag device to different signal acquisition devices based on the same service requirement. The signal acquisition device can be a UWB positioning base station.

[0119] For example, at the 1st second, the original user tag device 1 sends positioning signal 1 to 3 signal acquisition devices (the number of positioning signal 1 is 3), and at the 3rd second, the original user tag device 1 sends positioning signal 3 to 3 signal acquisition devices (the number of positioning signal 3 is 3). Then, the device identifiers of the original user device tags of positioning signal 1 and positioning signal 3 are the same, but the signal flag bits of positioning signal 1 and positioning signal 3 are different, and among the 3 positioning signal 1 (or 3 positioning signal 3), the reception timestamps are also different.

[0120] Multiple signal acquisition devices are installed in a scene area, which includes multiple unit scene areas, and the detection area is one of the multiple unit scene areas.

[0121] The server divides these multiple positioning signals into multiple positioning signal sets according to the signal flag bits of each positioning signal and the above detection time period. The signal flag bits in each positioning signal set are the same, the device identifiers of the original user label devices of the positioning signals in each positioning signal set are the same, the number of positioning signals included in each positioning signal set is the same and this number is the same as the number of signal acquisition devices, and the average reception timestamps of all the positioning signals in each positioning signal set are within the detection time period. There may be positioning signal sets with the same original user label device among these multiple positioning signal sets because the original label device continuously sends positioning signals to the signal acquisition devices. Therefore, within one detection time period, the same original label device may send multiple positioning signals to the signal acquisition devices.

[0122] The server determines the positioning coordinates of the original user label device corresponding to each positioning signal set according to each positioning signal set, and combines the determined positioning coordinates of the multiple original user label devices and the device identifiers of the multiple original user label devices into multiple label positioning information within this detection time period. Further, the average reception timestamp of the positioning signal set is used as the positioning timestamp of this label positioning information. It can be known that each label positioning information within the detection time period includes the device identifier of the original user label device and the positioning coordinates of the original user label device, and may also include the positioning timestamp.

[0123] For any one of the multiple positioning signal sets, the process of determining the positioning coordinates of the original user label device corresponding to this any one positioning signal set is as follows:

[0124] The server obtains the device coordinates of each signal acquisition device, obtains the reception timestamp of each positioning signal in this any one positioning signal set, and determines the positioning coordinates of the original label device of this any one positioning signal set according to the device coordinates and the reception timestamp of each positioning signal in this any one positioning signal set.

[0125] Taking the positioning signal set including 4 positioning signals as an example below, it shows how to determine the positioning coordinates corresponding to this positioning signal set: Please refer to Figure 5b , which is a schematic diagram for determining the positioning coordinates of a user label device provided in an embodiment of the present application. The 4 positioning signals can correspond to 4 signal acquisition devices. The server can use the following formula (1) to determine the positioning coordinates of a user label device:

[0126]

[0127] Among them, t1, t2, t3, and t4 respectively represent the reception timestamps of the positioning signals collected by the first signal acquisition device, the reception timestamps of the positioning signals collected by the second signal acquisition device, the reception timestamps of the positioning signals collected by the third signal acquisition device, and the reception timestamps of the positioning signals collected by the fourth signal acquisition device. c represents the propagation speed of the signal in the air. (x 1 , y 1 , z 1 ) represents the device coordinates collected by the first signal acquisition device, (x 2 , y 2 , z 2 ) represents the device coordinates collected by the second signal acquisition device, (x 3 , y 3 , z 3 ) represents the device coordinates collected by the third signal acquisition device, (x 4 , y 4 , z 4 ) represents the device coordinates collected by the fourth signal acquisition device, (x i , y i , z i ) represents the positioning coordinates of the original user tag device.

[0128] Step S303: Among the multiple positioning coordinates, use the positioning coordinates within the detection area as the target positioning coordinates.

[0129] Specifically, at this point, the server has obtained multiple tag positioning information during the detection time period, and each tag positioning information includes the device identifier of the original user tag device and the positioning coordinates of the original user tag device. Among the multiple positioning coordinates, use the positioning coordinates within the detection area as the target positioning coordinates.

[0130] Step S304: Use the device identifier of the original user tag device corresponding to the target positioning coordinates as the device identifier of the user tag device located in the detection area during the detection time period.

[0131] Specifically, the server refers to the original user tag device corresponding to the target positioning coordinates as the user tag device located in the detection area during the detection time period, and uses the device identifier of the original user tag device corresponding to the target positioning coordinates as the device identifier of the user tag device located in the detection area during the detection time period.

[0132] Step S305: Determine the reference user identity corresponding to the device identifier of the user tag device.

[0133] Specifically, the server determines the user identity corresponding to the device identifier of the user tag device obtained by querying the database (referred to as the reference user identity).

[0134] As can be seen from the above, this application uses the TDOA positioning method in UWB positioning technology to determine the positioning coordinates of each user tag device. The TDOA positioning method is a high-precision positioning method with a positioning error within 1 meter, which can ensure the accuracy of the identified positioning coordinates. Furthermore, it can accurately judge the situation of human-card separation and send an alarm message for human-card separation precisely.

[0135] Please refer to Figure 6 , which is an interaction diagram for detecting human-card separation provided by an embodiment of this application. Determining human-card separation involves the work card, the positioning platform, and the vision platform. Among them, the positioning platform and the vision platform can correspond to the Figures 3 - 5b server in the corresponding embodiment. The following embodiments mainly describe that the vision platform actively requests positioning information from the positioning platform, and then the vision platform completes the alarm judgment. The specific process of detecting human-card separation is as follows:

[0136] Step S401, a person wears a work card and enters the IDC (Internet Data Center) computer room. The work card sends UWB positioning signals around the computer room, and these signals are received and measured by UWB positioning base stations, and each generates measurement information. The measurement information at least includes: arrival timestamp and work card identifier.

[0137] The work card identifier can correspond to the device identifier of the original user identification device in this application, the arrival timestamp can correspond to the reception timestamp in this application, and the measurement information can correspond to the positioning signal in this application.

[0138] Step S402, the positioning platform receives the measurement information sent by the positioning base station and determines the position of the work card.

[0139] Specifically, the positioning platform is connected to the UWB positioning base station, and the UWB base station sends the above measurement information to the positioning platform. The positioning platform takes the base station coordinates and the above positioning measurement information as inputs to calculate the position of the work card. The UWB positioning accuracy is within 1 meter. And the positioning platform determines the position of the work card based on the TDOA positioning method in UWB positioning technology. Among them, the position of the work card can correspond to the positioning coordinates of the original user tag device in this application.

[0140] Step S403, the vision platform performs human detection, face detection, and work card detection on the video stream.

[0141] Specifically, the vision platform is connected to the camera and receives the video stream from the camera. The vision platform processes the received video stream for human detection, face recognition, and work card detection. Among them, human detection is to detect whether there are humans and their numbers, face recognition is to detect faces and identify their identities, and work card detection is to detect whether the person is wearing a work card. Under normal circumstances, the number of human detections is equal to the number of faces, which is equal to the number of work cards. However, simply using the vision method cannot determine whether there are security incidents such as work card borrowing, and the positioning ability of the positioning platform is also required.

[0142] Step S404, the vision platform requests work card positioning information from the positioning platform.

[0143] Specifically, the vision platform sends a request message for obtaining work card positioning information to the positioning platform. The request message is used to request the work card situation within a specified area and within a specified time period from the positioning platform. The request message at least includes the following information: 1) area number, 2) time period information. The area number and the time period information instruct the positioning platform to return the work card information for the specified area and the specified time period.

[0144] The area corresponding to the area number can correspond to the detection area in this application, and the time period information can correspond to the detection time period in this application. Moreover, the area included in the video stream collected by the vision platform contains the area corresponding to the area number, and the start collection time and the end collection time of the video stream collected by the vision platform belong to the time period information.

[0145] Step S405, after receiving the above request message, the positioning platform sends a work card response message for the specified area and the specified time period to the vision platform.

[0146] Specifically, the work card response message at least includes: 1) area number, 2) number of work cards, 3) work card identifier and its location, 4) time period information. Among them, the work card identifier corresponds to the identity or category of the person. For example, the work card identifier of an employee is in one-to-one correspondence with the identity of the employee, and one cannot wear someone else's work card.

[0147] Among them, the number of work cards can correspond to the number of reference user identities in this application.

[0148] Step S406, the vision platform combines the vision processing results and the work card positioning information to complete the person-card separation detection.

[0149] Specifically, there are several typical situations for human-card separation detection: 1) Someone is not wearing a card. For example, the vision platform identifies 5 people, but the positioning platform only locates the positions of 4 work cards. By comparing the number of people in the area, it can be determined that one person is not wearing a work card, and an alarm can be issued. 2) The work card worn is not of the person's own. The vision platform identifies the face of the person corresponding to the area number in the area, and then identifies their identity. Then, it checks the work card position information. If it is detected that the work cards corresponding to all the faces are also in the area corresponding to the area number, it means it is normal (that is, the person entering the IDC has passed the identity verification); if the work card corresponding to the face is not in the corresponding area, it means there is a situation of human-card separation (that is, the person entering the IDC has not passed the identity verification), and an alarm can be issued. Generally speaking, the vision platform determines whether there is a human-card separation event by comparing the processing results of the vision platform and the positioning information of the positioning platform.

[0150] Please refer to Figure 7 , which is an interaction diagram for detecting human-card separation provided by an embodiment of the present application. Determining human-card classification involves work cards, a positioning platform, and a vision platform, where the positioning platform and the vision platform can correspond to the Figures 3 - 5b corresponding servers in the foregoing embodiments. The following embodiments mainly describe that the positioning platform actively requests the vision processing results from the vision platform, and then the positioning platform completes the alarm judgment. The specific process of detecting human-card separation is as follows:

[0151] Step S501, a person wears a work card and enters the IDC (Internet Data Center) computer room. The work card sends out UWB positioning signals around the computer room, and these signals are received and measured by UWB positioning base stations, and each generates measurement information. This measurement information at least includes: arrival timestamp and work card identification.

[0152] Step S502, the positioning platform receives the measurement information sent by the positioning base station and determines the work card position of the work card.

[0153] Step S503, the vision platform performs human body detection, face detection, and work card detection on the video stream.

[0154] Among them, the specific processes of steps S501 - S503 can refer to steps S401 - S403 in the Figure 6 corresponding embodiments above.

[0155] Step S504, the positioning platform sends a request message to the vision platform to request the human body detection result and the face recognition result.

[0156] Specifically, the positioning platform sends a request message to the vision platform to obtain the human detection result and the face recognition result. The request message is used to request the vision platform for the video recognition situation within a specified area and within a specified time period. The request message includes at least the following information: 1) area number, 2) time period information. The area number and the time period information instruct the vision platform to return the human detection result and the face recognition result for the specified area and the specified time period. The area corresponding to the area number may correspond to the detection area in this application, and the time period information may correspond to the detection time period in this application. Moreover, the area corresponding to the area number is included in the area where the work card position determined by the positioning platform is located, and the positioning timestamp of the work card position determined by the positioning platform belongs to the time period information.

[0157] Step S505, after receiving the above request message, the vision platform sends a video detection response message for the specified area and the specified time period to the positioning platform.

[0158] Specifically, the video detection response message includes at least: 1) area number, 2) number of people, 3) identity of people, 4) time period information. Among them, the identity of people corresponds to the work card identifier. For example, the work card identifier of an employee and the identity of this employee are in one-to-one correspondence, and it is not allowed to wear someone else's work card.

[0159] Step S506, the positioning platform combines the vision processing result and the work card positioning information to complete the detection of work card and person separation.

[0160] Specifically, there are several typical situations for the detection of work card and person separation: 1) Someone is without a work card. For example, the vision platform identifies 5 people, but the positioning platform only locates 4 work card positions. By comparing the number of people in the area, it can be judged that one person is without a work card, and an alarm can be issued. 2) The work card worn is not of the person's own. The positioning platform detects whether the work card corresponding to the identity of the person returned by the vision platform is in the area corresponding to the area number. If it is, it means it is normal; if not, it means there is a situation of work card and person separation, and an alarm can be issued. Generally speaking, the positioning platform determines whether there is an event of work card and person separation by comparing the processing result of the vision platform and the positioning information of the positioning platform.

[0161] Please refer to Figure 8 , which is an interaction diagram for detecting work card and person separation provided by an embodiment of this application. Determining work card and person separation involves work cards, positioning platforms, and vision platforms. Among them, the positioning platform and the vision platform can correspond to the Figures 3 - 5b corresponding servers in the foregoing embodiments. The following embodiments mainly describe that the positioning platform actively synchronizes information to the vision platform in real time, and then the vision platform completes the alarm judgment. The specific process of detecting work card and person separation is as follows:

[0162] Step S601, a person wears a work card and enters the IDC (Internet Data Center) computer room. The work card sends UWB positioning signals around the computer room. These signals are received and measured by UWB positioning base stations, and each generates measurement information. The measurement information at least includes: arrival timestamp and work card identification.

[0163] Step S602, the positioning platform receives the measurement information sent by the positioning base station and determines the location of the work card.

[0164] Step S603, the vision platform performs human body detection, face detection, and work card detection on the video stream.

[0165] Among them, the specific processes of steps S601 - S603 can refer to steps S401 - S403 in the corresponding Figure 6 embodiment.

[0166] Step S604, the positioning platform actively synchronously sends a work card response message for a specified area and a specified time period to the vision platform.

[0167] Specifically, without the vision platform requesting, the latency can be reduced. The work card response message at least includes: 1) area number, 2) number of work cards, 3) work card identification and its location, 4) time period information. Among them, the work card identification corresponds to the identity or category of the person. For example, the work card identification of an employee corresponds one-to-one with the identity of that employee, and one cannot wear another person's work card.

[0168] Step S605, the vision platform combines the vision processing results and the work card positioning information to complete the detection of person-card separation.

[0169] Among them, the specific process of step S605 can refer to Figure 6 step S406 in the corresponding embodiment.

[0170] Please refer to Figure 9 , which is an interaction diagram for detecting person-card separation provided by an embodiment of the present application. Determining person-card classification involves work cards, a positioning platform, and a vision platform. Among them, the positioning platform and the vision platform can correspond to the Figures 3 - 5b server in the corresponding embodiment. The following embodiments mainly describe that the vision platform actively synchronously sends information to the positioning platform in real time, and then the positioning platform completes the alarm judgment. The specific process of detecting person-card separation is as follows:

[0171] Step S701, a person wears a work card and enters the IDC (Internet Data Center) computer room. The work card sends UWB positioning signals around the computer room. These signals are received and measured by UWB positioning base stations, and each generates measurement information. The measurement information at least includes: arrival timestamp and work card identification.

[0172] Step S702: The positioning platform receives the measurement information sent by the positioning base station and determines the position of the work card.

[0173] Step S703: The vision platform performs human detection, face detection, and work card detection on the video stream.

[0174] Among them, the specific processes of steps S701 - S703 can refer to steps S401 - S403 in the corresponding Figure 6 embodiment.

[0175] Step S704: The vision platform actively synchronously sends a video detection response message for a specified area and a specified time period to the positioning platform.

[0176] Specifically, without the positioning platform requesting, the time delay can be shortened. The video detection response message at least includes: 1) area number, 2) number of personnel, 3) identity of personnel, 4) time period information. Among them, the identity of personnel corresponds to the work card identification. For example, the work card identification of an employee and the identity of this employee are in one-to-one correspondence, and one cannot wear another person's work card.

[0177] Step S705: The positioning platform combines the vision processing result and the work card positioning information to complete the person-card separation detection.

[0178] Among them, the specific process of step S705 can refer to Figure 7 step S506 in the corresponding embodiment.

[0179] Further, please refer to Figure 10 , which is a schematic structural diagram of a data processing device 1 provided in an embodiment of the present application. The data processing device 1 can be applied to the server in the corresponding Figures 3 - 5b embodiment, as well as Figures 6 - 9 the positioning platform and the vision platform in the corresponding embodiment. Specifically, the data processing device 1 can be a computer program (including program code) running in a computer device.

[0180] The data processing device 1 may include: a first acquisition module 11, a second acquisition module 12, an identification module 13, and a first determination module 14.

[0181] The first acquisition module 11 is used to acquire a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period;

[0182] The second acquisition module 12 is used to acquire a video stream including a user to be verified, where the user to be verified refers to a user waiting for identity verification who enters the detection area during the detection time period;

[0183] An identification module 13, configured to determine the relationship between the user to be verified and the user tagging device according to the reference user identity and the video stream;

[0184] A first determination module 14, configured to determine the authentication result of the user to be verified according to the relationship between the user to be verified and the user tagging device.

[0185] In one embodiment, when the identification module 13 is configured to determine the relationship between the user to be verified and the user tagging device according to the reference user identity and the video stream, it is specifically configured to:

[0186] Perform video recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified; the information of the user to be verified includes at least one of the identity of the user to be verified of the user to be verified and the number of users of the user to be verified;

[0187] Determine the relationship between the user to be verified and the user tagging device according to the information of the user to be verified and the reference user identity, and the relationship between the user to be verified and the user tagging device includes a matching relationship or a non-matching relationship.

[0188] In one embodiment, the information of the user to be verified includes the identity of the user to be verified of the user to be verified, and the identity of the user to be verified is generated after performing face recognition processing on the video stream;

[0189] When the identification module 13 is configured to determine the relationship between the user to be verified and the user tagging device according to the information of the user to be verified and the reference user identity, it is specifically configured to:

[0190] If the reference user identity is the same as the identity of the user to be verified, determine that the relationship between the user to be verified and the user tagging device is a matching relationship; or,

[0191] If the reference user identity is different from the identity of the user to be verified, determine that the relationship between the user to be verified and the user tagging device is a non-matching relationship.

[0192] In one embodiment, the data processing device may further include: a second determination module 15.

[0193] The second determination module 15 is configured to obtain the number M of the identities of the user to be verified, and obtain the number N of the reference user identities; both M and N are positive integers;

[0194] The second determination module is further configured to determine that the reference user identities are different from the user identities to be verified if the number N of reference user identities is not equal to the number M of user identities to be verified; or, to determine that the reference user identities are different from the user identities to be verified if the number N of reference user identities is equal to the number M of user identities to be verified and there is no one-to-one correspondence between the N reference user identities and the M user identities to be verified; or, to determine that the reference user identities are the same as the user identities to be verified if the number N of reference user identities is equal to the number M of user identities to be verified and there is a one-to-one correspondence between the N reference user identities and the M user identities to be verified.

[0195] In one implementation, the user information to be verified includes the number of users to be verified, and the number of users is generated after performing human body recognition processing on the video stream;

[0196] When the recognition module 13 is used to determine the relationship between the user to be verified and the user label device according to the user information to be verified and the reference user identities, it is specifically configured to:

[0197] Obtain the number of reference user identities;

[0198] If the number of users to be verified is not equal to the number of reference user identities, determine that the relationship between the user to be verified and the user label device is a mismatch relationship.

[0199] In one implementation, when the first determination module 14 is used to determine the authentication result of the user to be verified according to the relationship between the user to be verified and the user label device, it is specifically configured to:

[0200] If the relationship between the user to be verified and the user label device is a match relationship, determine that the authentication result of the user to be verified is authentication passed; or,

[0201] If the relationship between the user to be verified and the user label device is a mismatch relationship, determine that the authentication result of the user to be verified is authentication failed.

[0202] In one implementation, when the first acquisition module 11 is used to acquire the reference user identities, it is specifically configured to:

[0203] Obtain a label device reporting request, where the label device reporting request includes a detection time period and a detection area;

[0204] Obtain multiple label positioning information within the detection time period, and each label positioning information includes the device identifier of the original user label device and the positioning coordinates of the original user label device;

[0205] Among the multiple positioning coordinates, use the positioning coordinates within the detection area as the target positioning coordinates;

[0206] Use the device identifier of the original user label device corresponding to the target positioning coordinate as the device identifier of the user label device located in the detection area during the detection time period;

[0207] Determine the reference user identity corresponding to the device identifier of the user label device.

[0208] In one implementation, when the first acquisition module 11 is used to acquire multiple tag positioning information during the detection time period, it is specifically used for:

[0209] Obtain multiple positioning signal sets from multiple signal acquisition devices. Any positioning signal set includes multiple positioning signals. Each positioning signal is sent by an original user label device. The device identifiers of the original user label devices in any positioning signal set are the same, and the average reception timestamp of any positioning signal set is within the detection time period;

[0210] According to each positioning signal set, determine the positioning coordinates of the original user label device corresponding to each positioning signal set;

[0211] Combine the positioning coordinates of multiple original user label devices and the device identifiers of multiple original user label devices into multiple tag positioning information during the detection time period.

[0212] In one implementation, for any positioning signal set among multiple positioning signal sets, when the first acquisition module 11 is used to determine the positioning coordinates of the original user label device corresponding to any positioning signal set according to any positioning signal set, it is specifically used for:

[0213] Obtain the device coordinates of multiple signal acquisition devices;

[0214] Obtain the reception timestamp of each positioning signal in any positioning signal set;

[0215] According to the device coordinates and the reception timestamp of each positioning signal in any positioning signal set, determine the positioning coordinates of the original user label device corresponding to any positioning signal set.

[0216] According to an embodiment of the present invention, Figures 3 - 5b Each step involved in the method shown can be performed by Figure 10 each module in the data processing device shown. For example, Figure 3 the steps S101 - S104 shown in can be performed by Figure 10 the first acquisition module 11, the second acquisition module 12, the identification module 13, and the first determination module 14 shown in respectively; and again, Figure 4 the steps S203 - S204 shown in can be performed by Figure 10executed by the recognition module 13 shown in Figure 5a The steps S301 - S305 shown in Figure 10 can be executed by the first acquisition module 11 shown in

[0217] Further, please refer to Figure 11 , Figure 11 which is a schematic structural diagram of a computer device provided by an embodiment of the present application. The above Figures 1 - 5b The server in the corresponding embodiment can be Figure 11 the computer device shown. As Figure 11 shown, the computer device may include: a processor 601, an input device 602, an output device 603, and a memory 604. The above processor 601, input device 602, output device 603, and memory 604 are connected through a bus 605. The memory 604 is used to store computer programs, and the computer programs include program instructions. The processor 601 is used to execute the program instructions stored in the memory 604.

[0218] In the embodiment of the present application, the processor 601 executes the following steps by running the executable program code in the memory 604:

[0219] Obtain a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period;

[0220] Obtain a video stream including the user to be verified, where the user to be verified refers to the user waiting for identity verification who enters the detection area during the detection time period;

[0221] Determine the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream;

[0222] Determine the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tag device.

[0223] In one embodiment, when the processor 601 executes to determine the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream, it specifically executes the following steps:

[0224] Perform video recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified; the information of the user to be verified includes at least one of the identity of the user to be verified and the number of users of the user to be verified;

[0225] Determine the relationship between the user to be verified and the user tag device according to the information of the user to be verified and the reference user identity, and the relationship between the user to be verified and the user tag device includes a matching relationship or a non - matching relationship.

[0226] In one embodiment, the user information to be verified includes the identity of the user to be verified, and the identity of the user to be verified is generated after performing face recognition processing on the video stream;

[0227] When the processor 601 determines the relationship between the user to be verified and the user tagging device according to the user information to be verified and the reference user identity, it specifically performs the following steps:

[0228] If the reference user identity is the same as the identity of the user to be verified, it is determined that the relationship between the user to be verified and the user tagging device is a matching relationship; or,

[0229] If the reference user identity is different from the identity of the user to be verified, it is determined that the relationship between the user to be verified and the user tagging device is a non-matching relationship.

[0230] In one embodiment, the processor 601 also performs the following steps:

[0231] Obtain the number M of the identities of the users to be verified, and obtain the number N of the reference user identities; both M and N are positive integers;

[0232] If the number N of the reference user identities is not equal to the number M of the identities of the users to be verified, it is determined that the reference user identity is different from the identity of the user to be verified; or,

[0233] If the number N of the reference user identities is equal to the number M of the identities of the users to be verified, and there is no one-to-one correspondence between the N reference user identities and the M identities of the users to be verified, it is determined that the reference user identity is different from the identity of the user to be verified; or,

[0234] If the number N of the reference user identities is equal to the number M of the identities of the users to be verified, and there is a one-to-one correspondence between the N reference user identities and the M identities of the users to be verified, it is determined that the reference user identity is the same as the identity of the user to be verified.

[0235] In one embodiment, the user information to be verified includes the number of users of the user to be verified, and the number of users is generated after performing human body recognition processing on the video stream;

[0236] When the processor 601 determines the relationship between the user to be verified and the user tagging device according to the user information to be verified and the reference user identity, it specifically performs the following steps:

[0237] Obtain the number of reference user identities;

[0238] If the number of users of the user to be verified is not equal to the number of reference user identities, it is determined that the relationship between the user to be verified and the user tagging device is a non-matching relationship.

[0239] In one embodiment, when the processor 601 executes to determine the authentication result of the user to be authenticated according to the relationship between the user to be authenticated and the user label device, the following steps are specifically executed:

[0240] Obtain the number of reference user identities;

[0241] If the relationship between the user to be authenticated and the user label device is a matching relationship, determine that the authentication result of the user to be authenticated is authentication passed; or,

[0242] If the relationship between the user to be authenticated and the user label device is a non-matching relationship, determine that the authentication result of the user to be authenticated is authentication failed.

[0243] In one embodiment, when the processor 601 executes to obtain the reference user identity, the following steps are specifically executed:

[0244] Obtain the label device reporting request, where the label device reporting request includes the detection time period and the detection area;

[0245] Obtain multiple label positioning information within the detection time period, and each label positioning information includes the device identifier of the original user label device and the positioning coordinates of the original user label device;

[0246] Among the multiple positioning coordinates, use the positioning coordinates within the detection area as the target positioning coordinates;

[0247] Use the device identifier of the original user label device corresponding to the target positioning coordinates as the device identifier of the user label device located in the detection area within the detection time period;

[0248] Determine the reference user identity corresponding to the device identifier of the user label device.

[0249] In one embodiment, when the processor 601 executes to obtain multiple label positioning information within the detection time period, the following steps are specifically executed:

[0250] Obtain multiple positioning signal sets from multiple signal acquisition devices. Any positioning signal set includes multiple positioning signals, each positioning signal is sent by an original user label device, the device identifiers of the original user label devices in any positioning signal set are the same, and the average reception timestamp of any positioning signal set is within the detection time period;

[0251] According to each positioning signal set, determine the positioning coordinates of the original user label device corresponding to each positioning signal set;

[0252] Combine the positioning coordinates of multiple original user label devices and the device identifiers of multiple original user label devices into multiple label positioning information within the detection time period.

[0253] In one embodiment, for any one of a plurality of positioning signal sets, when the processor 601 executes to determine the positioning coordinates of the original user tag device corresponding to any one of the positioning signal sets according to any one of the positioning signal sets, the following steps are specifically executed:

[0254] Obtain the device coordinates of a plurality of signal acquisition devices;

[0255] Obtain the reception timestamp of each positioning signal in any one of the positioning signal sets;

[0256] Determine the positioning coordinates of the original user tag device corresponding to any one of the positioning signal sets according to the device coordinates and the reception timestamp of each positioning signal in any one of the positioning signal sets.

[0257] It should be understood that the computer device 1000 described in the embodiments of the present application can execute the description of the data processing method in the corresponding embodiments mentioned above, and can also execute the description of the data processing device 1 in the corresponding embodiments mentioned above, which will not be elaborated here. In addition, the description of the beneficial effects of adopting the same method will not be elaborated either. Figures 3 - 9 The description of the data processing method in the corresponding embodiments mentioned above, and can also execute the description of the data processing device 1 in the corresponding embodiments mentioned above, which will not be elaborated here. In addition, the description of the beneficial effects of adopting the same method will not be elaborated either. Figure 10 For the beneficial effects of adopting the same method, no further elaboration will be made.

[0258] In addition, it should be noted here that: the embodiments of the present application also provide a computer storage medium, and the computer storage medium stores a computer program executed by the data processing device 1 mentioned above, and the computer program includes program instructions. When the processor executes the program instructions, it can execute the description of the data processing method in the corresponding embodiments mentioned above. Therefore, it will not be elaborated here. In addition, the description of the beneficial effects of adopting the same method will not be elaborated either. For the technical details not disclosed in the embodiments of the computer storage medium involved in the present application, please refer to the description of the method embodiments of the present application. Figures 3 - 9 For the beneficial effects of adopting the same method, no further elaboration will be made. For the technical details not disclosed in the embodiments of the computer storage medium involved in the present application, please refer to the description of the method embodiments of the present application.

[0259] According to one aspect of the present application, there is provided a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device can execute the method in the corresponding embodiments mentioned above. Therefore, it will not be elaborated here. Figures 3 to 9 For the beneficial effects of adopting the same method, no further elaboration will be made.

[0260] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The above program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above various methods. Among them, the above storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0261] The above-disclosed are only the preferred embodiments of the present application. Of course, the scope of rights of the present application cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A data processing method, characterized in that, it includes: Obtain a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period; Obtain a video stream including the user to be verified, where the user to be verified refers to a user waiting for identity verification who enters the detection area during the detection time period; Determine the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream; Determine the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tag device; Wherein, the determining the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream includes: When the light of the video stream is dim, perform human body recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified; the information of the user to be verified includes the number of users of the user to be verified; When the light of the video stream is bright, perform video recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified; the information of the user to be verified includes at least one of the identity of the user to be verified of the user to be verified and the number of users of the user to be verified; the video recognition processing includes at least one of face recognition processing and human body recognition processing; Determine the relationship between the user to be verified and the user tag device according to the information of the user to be verified and the reference user identity, and the relationship between the user to be verified and the user tag device includes a matching relationship or a non-matching relationship; Wherein, the human body recognition processing includes: determining the human body area in each video frame image in the video stream, counting the number of areas of the human body area in each video frame image, and taking the average value of the number of areas of the human body areas in all video frame images in the video stream as the number of users of the user to be verified.

2. The method according to claim 1, characterized in that, the information of the user to be verified includes the identity of the user to be verified of the user to be verified, and the identity of the user to be verified is generated after performing face recognition processing on the video stream; the determining the relationship between the user to be verified and the user tag device according to the information of the user to be verified and the reference user identity includes: If the reference user identity is the same as the identity of the user to be verified, determine that the relationship between the user to be verified and the user tag device is a matching relationship; Or, If the reference user identity is different from the identity of the user to be verified, determine that the relationship between the user to be verified and the user tag device is a non-matching relationship.

3. The method according to claim 2, characterized in that, the method further includes: Obtain the number M of the identity of the user to be verified, and obtain the number N of the reference user identity; both M and N are positive integers; If the number N of the reference user identity is not equal to the number M of the identity of the user to be verified, determine that the reference user identity is different from the identity of the user to be verified; or, If the number N of the reference user identities is equal to the number M of the user identities to be verified, and there is no one-to-one correspondence between the N reference user identities and the M user identities to be verified, it is determined that the reference user identity is different from the user identity to be verified; or, If the number N of the reference user identities is equal to the number M of the user identities to be verified, and there is a one-to-one correspondence between the N reference user identities and the M user identities to be verified, it is determined that the reference user identity is the same as the user identity to be verified.

4. The method according to claim 1, characterized in that, the user information to be verified includes the number of users of the user to be verified, and the number of users is generated after performing human body recognition processing on the video stream; determining the relationship between the user to be verified and the user tagging device according to the user information to be verified and the reference user identity includes: obtaining the number of the reference user identities; If the number of users of the user to be verified is not equal to the number of the reference user identities, it is determined that the relationship between the user to be verified and the user tagging device is a mismatch relationship.

5. The method according to any one of claims 1-4, characterized in that, determining the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tagging device includes: If the relationship between the user to be verified and the user tagging device is a matching relationship, it is determined that the identity verification result of the user to be verified is that the identity verification is passed; or, If the relationship between the user to be verified and the user tagging device is a mismatch relationship, it is determined that the identity verification result of the user to be verified is that the identity verification fails.

6. The method according to claim 1, characterized in that, obtaining the reference user identity includes: obtaining a reporting request from a tagging device, where the reporting request from the tagging device includes the detection time period and the detection area; obtaining a plurality of tag positioning information within the detection time period, and each tag positioning information includes the device identifier of the original user tagging device and the positioning coordinates of the original user tagging device; among the plurality of positioning coordinates, taking the positioning coordinates located in the detection area as the target positioning coordinates; taking the device identifier of the original user tagging device corresponding to the target positioning coordinates as the device identifier of the user tagging device located in the detection area within the detection time period; determining the reference user identity corresponding to the device identifier of the user tagging device.

7. The method according to claim 6, characterized in that, obtaining the plurality of tag positioning information within the detection time period includes: obtaining a plurality of positioning signal sets from a plurality of signal acquisition devices, where any positioning signal set includes a plurality of positioning signals, each positioning signal is sent by an original user tagging device, the device identifiers of the original user tagging devices of any positioning signal set are the same, and the average reception timestamp of any positioning signal set is within the detection time period; determining the positioning coordinates of the original user tagging device corresponding to each positioning signal set according to each positioning signal set; Combine the positioning coordinates of multiple original user tag devices and the device identifiers of multiple original user tag devices into multiple tag positioning information within the detection time period.

8. The method according to claim 7, wherein, for any positioning signal set among the multiple positioning signal sets, the process of determining the positioning coordinates of the original user tag device corresponding to the any positioning signal set according to the any positioning signal set includes: Obtain the device coordinates of the multiple signal acquisition devices; Obtain the reception timestamp of each positioning signal in the any positioning signal set; Determine the positioning coordinates of the original user tag device corresponding to the any positioning signal set according to the device coordinates and the reception timestamp of each positioning signal in the any positioning signal set.

9. A data processing device, wherein, comprising: A first acquisition module, configured to acquire a reference user identity, where the reference user identity refers to the user identity corresponding to the user tag device located in the detection area during the detection time period; A second acquisition module, configured to acquire a video stream including a user to be verified, where the user to be verified refers to a user waiting for identity verification who enters the detection area during the detection time period; An identification module, configured to determine the relationship between the user to be verified and the user tag device according to the reference user identity and the video stream; A first determination module, configured to determine the identity verification result of the user to be verified according to the relationship between the user to be verified and the user tag device; wherein, the identification module is specifically configured to: When the light of the video stream is dim, perform human body recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified; the information of the user to be verified includes the number of users of the user to be verified; When the light of the video stream is bright, perform video recognition processing on the video stream to obtain the information of the user to be verified of the user to be verified; the information of the user to be verified includes at least one of the identity of the user to be verified of the user to be verified and the number of users of the user to be verified; the video recognition processing includes at least one of face recognition processing and human body recognition processing; Determine the relationship between the user to be verified and the user tag device according to the information of the user to be verified and the reference user identity, and the relationship between the user to be verified and the user tag device includes a matching relationship or a non-matching relationship; wherein, the human body recognition processing includes: determining the human body area in each video frame image in the video stream, counting the number of areas of the human body area in each video frame image, and taking the average value of the number of areas of the human body area in all video frame images in the video stream as the number of users of the user to be verified.

10. A computer device, wherein, comprises a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1-8.

11. A computer storage medium, wherein, The computer storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, the method according to any one of claims 1-8 is executed.

Images