Method and system for realizing personnel passing management based on streaming media, and storage medium

By obtaining streaming media data and analyzing facial recognition technology, identifying unauthorized personnel and triggering early warning mechanisms, the problem of failure to effectively supervise unauthorized personnel in the existing technology is solved, and the efficiency of security management and community security is improved.

CN119007327APending Publication Date: 2024-11-22FUZHOU MILI TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411036099.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Existing facial recognition technology is difficult to effectively supervise unauthorized personnel, resulting in strangers breaking in to increase the probability of dangerous events.

Method used

By obtaining streaming media data, extracting face data and determining whether it is an unauthorized person, recording their behavioral trajectory, analyzing whether the warning conditions are met, if they are met, triggering the warning mechanism, including the first and second warning levels, increasing the monitoring frequency and analysis In-depth, notify security personnel in a timely manner.

Benefits of technology

Continuous behavior monitoring and early warning of unauthorized personnel has been achieved, safety management efficiency has been improved, necessary safety measures have been taken in a timely manner, and the probability of safety incidents has been reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119007327A_ABST
    Figure CN119007327A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of access control management, in particular to a method and system for realizing personnel passing management based on streaming media and a storage medium. The method comprises the following steps: acquiring streaming media data and extracting face data from the streaming media data; judging whether the person is an unauthorized person or not according to the extracted face data; and if yes, recording a behavior track of an unauthorized person and analyzing whether the behavior track meets an early warning condition, and if the behavior track meets the early warning condition, triggering an early warning mechanism. The method comprises the following steps of: continuously monitoring behaviors of unauthorized personnel, analyzing a moving path of the personnel in a monitoring area, identifying a behavior mode of the personnel, and if an early warning condition is met, triggering an early warning mechanism, for example, immediately pushing detailed alarm information to a management platform and quickly informing security personnel through multiple communication channels, so that the security personnel can be quickly informed. Security personnel can obtain alarm and evaluation conditions in time and take necessary safety measures, so that personnel passing management is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of access control management, and in particular to a method, system and storage medium for implementing personnel passage management based on streaming media. Background Art

[0002] With the development of biometrics, the Internet of Things, and cloud computing technologies, the intelligent access control management system has been continuously enriched, realizing the intelligent and remote management of building entrances and exits. However, it is also slightly insufficient in terms of convenience of passage, equipment compatibility, and system stability.

[0003] At present, communities usually use integrated identification machines as credential identification terminals, which mainly rely on face (encrypted card) recognition technology to quickly authenticate and identify community members' access rights. The application of this technology has enhanced the security of the community to a certain extent while improving the efficiency of community security management and reducing the risk of losing traditional keys or access cards. However, although this method has significant advantages, it also inevitably has some limitations and problems. Current face recognition devices cannot effectively supervise unauthorized persons, such as being unable to accurately identify incidents such as people following intrusions, resulting in strangers breaking in and increasing the probability of dangerous incidents. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a method, system and storage medium for implementing personnel passage management based on streaming media.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: A method for implementing personnel passage management based on streaming media includes the following steps: S1, obtaining streaming media data and extracting face data from the streaming media data; S2. Determine whether the person is an unauthorized person based on the extracted facial data; S3. If yes, the behavior track of the unauthorized person is recorded and analyzed to see whether the behavior track meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

[0006] Furthermore, it also includes: presetting an authorization database, wherein the authorization database is used to store facial data corresponding to authorized personnel; Step S3 includes: The extracted facial data is compared with the facial data corresponding to the authorized person in the authorization database. If the extracted facial data is of an authorized person, a door unlocking command is sent to the access control system; if the extracted facial data is of an unauthorized person, the behavior trajectory of the unauthorized person is recorded and analyzed to see whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

[0007] Furthermore, it also includes: presetting a behavior database, wherein the behavior database is used to store behavior trajectory data corresponding to abnormal behaviors; In step S3, analyzing whether the behavior trajectory meets the warning condition includes: The recorded behavior trajectory of the unauthorized person is compared with the behavior trajectory data in the behavior database. If the recorded behavior trajectory of the unauthorized person is determined to be abnormal behavior, the early warning mechanism is triggered.

[0008] Furthermore, step S3 also includes: The behavior trajectory of the unauthorized person is recorded until the facial data of the unauthorized person is no longer extracted from the streaming data and the recording is stopped.

[0009] Further, the warning condition in step S3 includes a first warning level and a second warning level, and the second warning level is higher than the first warning level; When the behavior trajectory meets the first warning level and does not meet the second warning level, increase the monitoring frequency and analysis depth of the unauthorized personnel; When the behavior trajectory meets the second warning level, the warning mechanism is triggered.

[0010] Furthermore, it also includes: A blacklist database is preset; the blacklist database is used to store facial data corresponding to abnormal persons; After step S3, the following steps are also included: The facial data corresponding to the unauthorized persons who triggered the early warning mechanism will be added to the access blacklist database.

[0011] Furthermore, step S2 further includes: Determine whether the facial data in the blacklist database contains the extracted facial data; If so, the early warning mechanism is triggered.

[0012] Furthermore, step S1 is specifically as follows: Acquire streaming media data, generate a surveillance camera continuous picture stream that supports standard video stream format output, and perform image enhancement processing on the surveillance camera continuous picture stream to obtain the surveillance camera continuous picture stream after image enhancement processing; Extract face data from the continuous image stream of surveillance cameras after image enhancement processing.

[0013] A system for implementing personnel access management based on streaming media includes a processor and a memory, wherein the memory stores a program or instruction, and when the program or instruction is executed by the processor, the method for implementing personnel access management based on streaming media is implemented.

[0014] A computer storage medium includes: a readable storage medium and a computer program, wherein the computer program is used to implement the method for realizing personnel passage management based on streaming media.

[0015] The beneficial effects of the present invention are: The present invention provides a method, system and storage medium for implementing personnel access management based on streaming media, which obtains streaming media data and extracts face data from the streaming media data; then determines whether it is an unauthorized person based on the extracted face data; if so, records the behavior trajectory of the unauthorized person and analyzes whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered. That is, there is continuous behavior monitoring of unauthorized personnel, the movement path of the person in the monitoring area is analyzed, and his behavior pattern is identified. If there is a warning condition that is met (such as following others to enter or attempting to force entry, etc.), the warning mechanism is triggered, such as immediately pushing detailed alarm information to the management platform, and quickly notifying security personnel through multiple communication channels, such as text messages, emails or dedicated applications, so that security personnel can receive alarms in time, evaluate the situation, and take necessary safety measures, such as on-site investigation, intervention or alarm, etc., so as to achieve personnel access management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a flow chart of the steps of a method for realizing personnel passage management based on streaming media of the present invention; Figure 2 The figure shows a structural block diagram of a system for realizing personnel passage management based on streaming media of the present invention; Description of Figure Numbers: 1-Processor; 2-Memory. DETAILED DESCRIPTION

[0017] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments: like Figure 1 As shown, the present invention provides a method for implementing personnel passage management based on streaming media, comprising the following steps: S1, obtaining streaming media data and extracting face data from the streaming media data; S2. Determine whether the person is an unauthorized person based on the extracted facial data; S3. If yes, the behavior track of the unauthorized person is recorded and analyzed to see whether the behavior track meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

[0018] From the above description, it can be seen that the present invention has the following beneficial effects: The present invention provides a method for implementing personnel access management based on streaming media, which obtains streaming media data and extracts face data from the streaming media data; then determines whether the person is an unauthorized person based on the extracted face data; if so, records the behavior trajectory of the unauthorized person and analyzes whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered. That is, there is continuous behavior monitoring of unauthorized persons, the movement path of the person in the monitoring area is analyzed, and the behavior pattern is identified. If the warning conditions are met (such as following others to enter or attempting to forcefully enter, etc.), the warning mechanism is triggered, such as immediately pushing detailed alarm information to the management platform, and quickly notifying security personnel through multiple communication channels, such as text messages, emails or special applications, so that the security personnel can receive the alarm in time, evaluate the situation, and take necessary safety measures, such as on-site investigation, intervention or alarm, so as to achieve personnel access management.

[0019] Furthermore, it also includes: presetting an authorization database, wherein the authorization database is used to store facial data corresponding to authorized personnel; Step S3 includes: The extracted facial data is compared with the facial data corresponding to the authorized person in the authorization database. If the extracted facial data is of an authorized person, a door unlocking command is sent to the access control system; if the extracted facial data is of an unauthorized person, the behavior trajectory of the unauthorized person is recorded and analyzed to see whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

[0020] From the above description, it can be seen that the present invention has the following beneficial effects: through the above method, it is possible to determine whether the extracted facial data is of an unauthorized person, specifically by performing a one-to-one comparison with the authorization database that stores the facial data corresponding to authorized persons, and obtaining the corresponding similarity value by calculating the similarity. If the similarity value is greater than a preset threshold, the comparison is considered successful. If the similarity value is not greater than the preset threshold, the comparison is considered failed, and the process is traversed in sequence until all comparisons are completed.

[0021] Furthermore, it also includes: presetting a behavior database, wherein the behavior database is used to store behavior trajectory data corresponding to abnormal behaviors; In step S3, analyzing whether the behavior trajectory meets the warning condition includes: The recorded behavior trajectory of the unauthorized person is compared with the behavior trajectory data in the behavior database. If the recorded behavior trajectory of the unauthorized person is determined to be abnormal behavior, the early warning mechanism is triggered.

[0022] From the above description, it can be seen that the present invention has the following beneficial effects: through the above method, the behavior trajectory of unauthorized persons can be judged, and when the recorded behavior trajectory of unauthorized persons is determined to be abnormal behavior, an early warning mechanism is triggered.

[0023] Furthermore, step S3 also includes: The behavior trajectory of the unauthorized person is recorded until the facial data of the unauthorized person is no longer extracted from the streaming data and the recording is stopped.

[0024] From the above description, it can be seen that the present invention has the following beneficial effects: through the above mechanism, when the facial data of unauthorized persons is not extracted from the streaming media data, recording is stopped, which greatly reduces unnecessary data processing, thereby greatly reducing the data processing pressure borne by the system.

[0025] Further, the warning condition in step S3 includes a first warning level and a second warning level, and the second warning level is higher than the first warning level; When the behavior trajectory meets the first warning level and does not meet the second warning level, increase the monitoring frequency and analysis depth of the unauthorized personnel; When the behavior trajectory meets the second warning level, the warning mechanism is triggered.

[0026] From the above description, it can be seen that the present invention has the following beneficial effects: by setting different degrees of warning levels, the warning level can be judged by the behavioral trajectory, and corresponding disposal measures can be executed according to the different degrees of warning levels. For example, when the first warning level is met and the second warning level is not met, the monitoring frequency and analysis depth of unauthorized personnel are increased; when the second warning level is met, the warning mechanism is triggered.

[0027] Furthermore, it also includes: A blacklist database is preset; the blacklist database is used to store facial data corresponding to abnormal persons; After step S3, the following steps are also included: The facial data corresponding to the unauthorized persons who triggered the early warning mechanism will be added to the access blacklist database.

[0028] From the above description, it can be seen that the present invention has the following beneficial effects: by setting up a blacklist database, the facial data corresponding to the unauthorized persons who trigger the early warning mechanism are added to the blacklist database, and key persons are monitored.

[0029] Furthermore, step S2 further includes: Determine whether the facial data in the blacklist database contains the extracted facial data; If so, the early warning mechanism is triggered.

[0030] From the above description, it can be seen that the present invention has the following beneficial effects: once a person in the access blacklist database is identified again, the person is directly monitored, which can simplify behavior collection and analysis and improve emergency response efficiency.

[0031] Furthermore, step S1 is specifically as follows: Acquire streaming media data, generate a surveillance camera continuous picture stream that supports standard video stream format output, and perform image enhancement processing on the surveillance camera continuous picture stream to obtain the surveillance camera continuous picture stream after image enhancement processing; Extract face data from the continuous image stream of surveillance cameras after image enhancement processing.

[0032] From the above description, it can be seen that the present invention has the following beneficial effects: through the above method, the continuous picture stream of the surveillance camera is subjected to image enhancement processing, which is beneficial to the subsequent rapid and accurate extraction of facial data.

[0033] like Figure 2 As shown, the present invention also provides a system for implementing personnel access management based on streaming media, including a processor 1 and a memory 2, wherein the memory 2 stores a program or instruction, and when the program or instruction is executed by the processor 1, the method for implementing personnel access management based on streaming media is implemented.

[0034] From the above description, it can be seen that the present invention has the following beneficial effects: The present invention provides a system for implementing personnel access management based on streaming media, which obtains streaming media data and extracts face data from the streaming media data; then determines whether the person is an unauthorized person based on the extracted face data; if so, records the behavior trajectory of the unauthorized person and analyzes whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered. That is, there is continuous behavior monitoring of unauthorized persons, the movement path of the person in the monitoring area is analyzed, and the behavior pattern is identified. If the warning conditions are met (such as following others to enter or attempting to forcefully enter, etc.), the warning mechanism is triggered, such as immediately pushing detailed alarm information to the management platform, and quickly notifying security personnel through multiple communication channels, such as text messages, emails or special applications, so that the security personnel can receive alarms in time, evaluate the situation, and take necessary safety measures, such as on-site investigation, intervention or alarm, so as to achieve personnel access management.

[0035] The present invention also provides a computer storage medium, comprising: a readable storage medium and a computer program, wherein the computer program is used to implement the method for realizing personnel passage management based on streaming media.

[0036] Several preferred embodiments are listed below to help those skilled in the art better understand the technical content of the present invention and the technical contribution made by the present invention relative to the prior art: Preferred embodiment one: The present invention provides a method for implementing personnel passage management based on streaming media, comprising the following steps: An authorization database is preset, and the authorization database is used to store facial data corresponding to authorized personnel; a behavior database is preset, and the behavior database is used to store behavior trajectory data corresponding to abnormal behavior; a blacklist database is preset; the blacklist database is used to store facial data corresponding to abnormal personnel; S1, obtaining streaming media data and extracting face data from the streaming media data; Step S1 is specifically as follows: Acquire streaming media data, generate a surveillance camera continuous picture stream that supports standard video stream format output, and perform image enhancement processing on the surveillance camera continuous picture stream to obtain the surveillance camera continuous picture stream after image enhancement processing; Extract face data from the continuous image stream of surveillance cameras after image enhancement processing.

[0037] S2. Determine whether the person is an unauthorized person based on the extracted facial data; Step S2 also includes: determining whether the face data in the blacklist database contains the extracted face data; if so, triggering the early warning mechanism. That is, once a person in the blacklist database is identified again, the person is directly monitored, which can simplify behavior collection and analysis and improve emergency response efficiency.

[0038] S3. If yes, the behavior track of the unauthorized person is recorded and analyzed to see whether the behavior track meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

[0039] The extracted facial data is compared with the facial data corresponding to the authorized person in the authorization database. If the extracted facial data is of an authorized person, the access control unlocking command is sent to the access control system; if the extracted facial data is of an unauthorized person, the behavior trajectory of the unauthorized person is recorded and analyzed to see whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered. In particular, to determine whether the extracted facial data is of an unauthorized person, a one-to-one comparison is performed with the authorization database storing the facial data corresponding to the authorized person, and the corresponding similarity value is obtained by calculating the similarity. If the similarity value is greater than the preset threshold, the comparison is considered successful. If the similarity value is not greater than the preset threshold, the comparison is considered failed, and the process is repeated until all comparisons are completed.

[0040] In step S3, analyzing whether the behavior trajectory meets the warning condition includes: The recorded behavior trajectory of the unauthorized person is compared with the behavior trajectory data in the behavior database. If the recorded behavior trajectory of the unauthorized person is determined to be abnormal behavior, the early warning mechanism is triggered. The behavior trajectory of the unauthorized person is recorded until the facial data of the unauthorized person is not extracted from the streaming data. Through the above mechanism, the recording is stopped when the facial data of the unauthorized person is not extracted from the streaming data, which greatly reduces unnecessary data processing, thus greatly reducing the data processing pressure borne by the system.

[0041] In step S3, the warning conditions include a first warning level and a second warning level, wherein the second warning level is higher than the first warning level; when the behavior trajectory meets the first warning level but does not meet the second warning level, the monitoring frequency and analysis depth of unauthorized personnel are increased; when the behavior trajectory meets the second warning level, the warning mechanism is triggered. By setting different levels of warning levels, the warning level can be determined by the behavior trajectory, and corresponding disposal measures can be performed according to different levels of warning levels. For example, when the first warning level is met but the second warning level is not met, the monitoring frequency and analysis depth of unauthorized personnel are increased; when the second warning level is met, the warning mechanism is triggered.

[0042] The facial data corresponding to the unauthorized personnel who triggered the early warning mechanism is added to the access blacklist database. By setting up the access blacklist database, the facial data corresponding to the unauthorized personnel who triggered the early warning mechanism is added to the access blacklist database, and key personnel are monitored.

[0043] Preferred embodiment 2: A system for implementing personnel access management based on streaming media includes a processor 1 and a memory 2, wherein the memory 2 stores a program or instruction, and when the program or instruction is executed by the processor 1, the method for implementing personnel access management based on streaming media is implemented.

[0044] Preferred embodiment three: The present invention provides a method for implementing personnel passage management based on streaming media, which specifically performs the following steps: 1. Video acquisition and processing Through real-time streaming technology, the terminal can receive and process continuous image streams from surveillance cameras that support standard video stream format output, and perform a series of image enhancement operations, including denoising, contrast adjustment, and sharpening, to optimize image quality, ensure the clarity of facial information and the accuracy of recognition, while also reducing the performance requirements for surveillance equipment.

[0045] By utilizing existing surveillance cameras and streaming media technology, the need to purchase and build additional hardware facilities is effectively reduced, thereby significantly reducing the overall cost of community construction and maintenance without sacrificing safety and efficiency.

[0046] 2. Face feature extraction First, the terminal will locally analyze and collect facial information from various angles in multiple streaming media files, and use the multi-dimensional integrated facial recognition algorithm to select the best facial information of the same person at the same time for data analysis and business judgment operations, without the need for passers-by to actively use specific postures or stop to cooperate with the equipment for facial recognition. Secondly, the terminal extracts key facial features of the face, such as facial contours, the position and ratio of the eyes, nose and mouth, etc. These features are then converted into digital form to form a unique facial signature for accurate comparison with the authorized personnel information stored in the local database of the terminal. The entire extraction process is analyzed by edge computing, which greatly reduces the pressure on the server.

[0047] By using multiple cameras to capture images from multiple angles and recognize faces in the picture, the impact of natural light shifting over time can be avoided, or the faces of short people such as children and wheelchair users can be better acquired without the need for active cooperation from the people, thus reducing behavioral interference with passers-by. Continuous learning and algorithm optimization can also be used to improve recognition accuracy and the effectiveness of the early warning mechanism.

[0048] 3. Video face comparison and event warning Authorized personnel access to the area: The system compares and confirms that the captured face matches the entry in the authorization database, that is, when the personnel identity information is successfully verified, the access control system will automatically respond, unlock the corresponding access control, and allow the person to enter. At the same time, the system will record the time, date and personnel information of this access event in detail, providing important data basis for daily inspections, subsequent audits, monitoring and data analysis.

[0049] Unauthorized personnel access the area: If the system compares and confirms that the personnel information has not been verified, that is, the face in the picture does not match the information in the authorization database, the access control system will remain locked, effectively preventing unauthorized personnel from entering. At this time, the system will continuously monitor the behavior of the person, use video analysis technology to continuously track and deeply analyze the actions of unauthorized personnel, analyze the movement path of the person in the monitoring area, and identify his behavior pattern until the person leaves the monitoring screen.

[0050] The system will analyze the person's behavior in real time and determine whether an early warning is needed. When suspicious behavior is identified, such as a person staying at the door for too long, the system will increase the monitoring frequency and analysis depth of the person; when a warning behavior such as a person trying to follow others in or trying to force entry is identified, the system will trigger the early warning mechanism. After the early warning is triggered, the system will immediately push detailed alarm information to the management platform and quickly notify security personnel through multiple communication channels such as SMS, email or dedicated applications. In this way, security personnel can receive the alarm in time, evaluate the situation, and take necessary safety measures such as on-site investigation, intervention or alarm. At the same time, the system will automatically add the person who triggered the early warning to the access blacklist and automatically deny their entry in future access attempts.

[0051] Blacklisted personnel access: If the system compares and confirms that a person is a blacklisted member, the access control system will remain locked, effectively preventing blacklisted personnel from entering. At the same time, the system will trigger an alarm to remind relevant personnel to focus on the area and the person's behavior.

[0052] In addition, the system will link all camera images in the management area to conduct real-time behavior analysis and location lock on blacklisted personnel. Through video analysis technology, the system can track the movement trajectory of the person in the monitoring area, evaluate his or her behavior pattern, and update his or her location information in real time until the person leaves the management area or the security personnel successfully handles the alarm. The combination of technology and manpower ensures that a quick and effective response can be made when facing potential threats.

[0053] Through visual behavior recognition technology, the terminal can evaluate and identify various behavior patterns and make judgments on different behaviors, thereby significantly improving the accuracy of the early warning mechanism and the safety of the management area.

[0054] 4. Equipment linkage The system can also improve the efficiency and convenience of personnel passage through equipment linkage. For example, after a person passes through the access control, the elevator is automatically sent to the corresponding floor; when the visitor is verified, the person being visited is notified that the visitor has arrived.

[0055] 5. Record traffic events In addition, the system will record and archive all face verification events in detail, including facial images, behavior patterns, and alarm responses of unauthorized persons. These records not only help with post-event review and analysis, but can also be used to train and optimize face recognition algorithms, improve the system's recognition accuracy and the effectiveness of the early warning mechanism. Through continuous learning and adaptation, the system can more intelligently identify potential security threats, improve the level of security management in communities or important areas, and ensure that it is both intelligent and efficient.

[0056] The present invention has been described by the above-mentioned relevant embodiments and drawings, but the above-mentioned embodiments are only examples for implementing the present invention. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, modifications and equivalent settings included in the spirit and scope of the claims are all included in the scope of the present invention.

Claims

1. A method for implementing personnel traffic management based on streaming media, characterized in that: The following steps are involved: S1, obtaining streaming media data and extracting face data from the streaming media data; S2. Determine whether the person is an unauthorized person based on the extracted facial data; S3. If yes, the behavior track of the unauthorized person is recorded and analyzed to see whether the behavior track meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

2. The method for implementing personnel traffic management based on streaming media according to claim 1 is characterized in that: Also includes: An authorization database is preset, and the authorization database is used to store facial data corresponding to authorized personnel; Step S3 includes: The extracted facial data is compared with the facial data corresponding to the authorized person in the authorization database. If the extracted facial data is of an authorized person, a door unlocking command is sent to the access control system; if the extracted facial data is of an unauthorized person, the behavior trajectory of the unauthorized person is recorded and analyzed to see whether the behavior trajectory meets the warning conditions. If the warning conditions are met, the warning mechanism is triggered.

3. The method for implementing personnel traffic management based on streaming media according to claim 2 is characterized in that: Also includes: A behavior database is preset, and the behavior database is used to store behavior trajectory data corresponding to abnormal behaviors; In step S3, analyzing whether the behavior trajectory meets the warning condition includes: The recorded behavior trajectory of the unauthorized person is compared with the behavior trajectory data in the behavior database. If the recorded behavior trajectory of the unauthorized person is determined to be abnormal behavior, the early warning mechanism is triggered.

4. The method for implementing personnel traffic management based on streaming media according to claim 1 is characterized in that: Step S3 also includes: The behavior trajectory of the unauthorized person is recorded until the facial data of the unauthorized person is no longer extracted from the streaming data and the recording is stopped.

5. The method for implementing personnel traffic management based on streaming media according to claim 1 is characterized in that: In step S3, the warning condition includes a first warning level and a second warning level, and the second warning level is higher than the first warning level; When the behavior trajectory meets the first warning level and does not meet the second warning level, increase the monitoring frequency and analysis depth of the unauthorized personnel; When the behavior trajectory meets the second warning level, the warning mechanism is triggered.

6. The method for implementing personnel traffic management based on streaming media according to claim 1 is characterized in that: Also includes: A blacklist database is preset; The blacklist database is used to store facial data corresponding to abnormal persons; After step S3, the following steps are also included: The facial data corresponding to the unauthorized persons who triggered the early warning mechanism will be added to the access blacklist database.

7. The method for implementing personnel traffic management based on streaming media according to claim 6 is characterized in that: Step S2 also includes: Determine whether the facial data in the blacklist database contains the extracted facial data; If so, the early warning mechanism is triggered.

8. The method for implementing personnel traffic management based on streaming media according to claim 1 is characterized in that: Step S1 is specifically as follows: Acquire streaming media data, generate a surveillance camera continuous picture stream that supports standard video stream format output, and perform image enhancement processing on the surveillance camera continuous picture stream to obtain the surveillance camera continuous picture stream after image enhancement processing; Extract face data from the continuous image stream of surveillance cameras after image enhancement processing.

9. A system for realizing personnel traffic management based on streaming media, characterized in that: It includes a processor and a memory, wherein the memory stores a program or instruction, and when the program or instruction is executed by the processor, the method for realizing personnel passage management based on streaming media as described in any one of claims 1 to 8 is implemented.

10. A computer storage medium, characterized in that include: A readable storage medium and a computer program, wherein the computer program is used to implement the method for achieving personnel passage management based on streaming media as described in any one of claims 1 to 8.

Citation Information

Patent Citations

  • Personnel recognition method and related product

    CN109767527A

  • Face recognition gate-free access control system and face recognition gate-free access control method

    CN111862430A

  • Entrance guard control method and device of intelligent antitheft door, electronic equipment and storage medium

    CN111951450A

  • Directional information acquisition method and system, computer equipment and storage medium

    CN113034763A

  • Video AI-based sensitive area personnel access efficient monitoring method

    CN117392607A