Park monitoring management system

Abstract

The application relates to the field of monitoring and monitoring technology, and provides a park monitoring management system, which comprises video monitoring equipment, alarm equipment and an application operation management platform, the application operation management platform comprises a management terminal, the alarm equipment and the video monitoring equipment are connected with the management terminal, each video monitoring equipment corresponds to one alarm equipment, a rule violation detection algorithm is arranged in the video monitoring equipment, the video monitoring equipment is used for generating alarm information when the rule violation detection algorithm detects that a video shot has a rule violation, and sending rule violation video frames corresponding to the rule violation and the alarm information to the management terminal, the management terminal is used for generating a prompt information after receiving the rule violation video frames and the alarm information, and generating alarm voice corresponding to the alarm information after receiving a confirmation alarm instruction of a user, and sending the alarm voice to the corresponding alarm equipment for voice broadcast, so that the purpose of timely alarming and stopping the rule violation is achieved.

Description

Technical Field

[0001] This invention relates to the field of monitoring and surveillance technology, and in particular to a park monitoring and management system. Background Technology

[0002] Currently, security monitoring in factories, warehouses, and other industrial parks relies on traditional video surveillance. This involves cameras monitoring various areas of the park in real-time and transmitting the footage to a control room. Staff in the control room then analyze the video to determine if any violations have occurred, such as factory workers not wearing reflective vests and / or safety helmets, or pedestrians entering driveways. Once a violation is identified, the control room staff notifies the administrator of the area where the violation occurred to conduct an on-site inspection and stop the violation.

[0003] In traditional park monitoring methods, manual judgment is required to determine whether a violation has occurred. This may lead to negligence by monitoring room staff, who may miss real-time video footage of a violation, or staff may lack experience and misjudge the violation. As a result, when a violation occurs in the park, it may not be alerted or stopped in time, thus causing a safety accident. Summary of the Invention

[0004] This invention provides a park monitoring and management system to solve the problem in the prior art that it is impossible to promptly alert and stop violations that occur in the park.

[0005] This invention provides a park monitoring and management system, including: video surveillance equipment, alarm equipment, and an application operation management platform. The application operation management platform includes: a management terminal; both the alarm equipment and the video surveillance equipment are connected to the management terminal, and each video surveillance equipment corresponds to one alarm equipment.

[0006] The video surveillance device is equipped with a violation detection algorithm. When the violation detection algorithm detects a violation in the captured video, the video surveillance device generates an alarm message and sends the violation video frame corresponding to the violation and the alarm message to the management terminal.

[0007] The management terminal is used to generate a reminder message after receiving a violation video frame and alarm information, and after receiving a user's confirmation alarm command, generate an alarm voice corresponding to the alarm information and send the alarm voice to the corresponding alarm device for voice broadcast.

[0008] According to a park monitoring and management system provided by the present invention, the video surveillance device is used to determine the current time as the alarm time and the location of the video surveillance device as the alarm location when the violation detection algorithm detects a violation in the captured video; determine the type of violation based on the output of the violation detection algorithm; determine the alarm level based on the type of violation or the number of violations detected by the same violation detection algorithm within a predetermined time period; and generate the alarm information based on the alarm time, alarm location, alarm level, and violation type.

[0009] According to a park monitoring and management system provided by the present invention, the management terminal is further configured to determine whether to generate a work order based on the alarm level. If it is determined that a work order should be generated, a work order is generated based on the alarm information, and the work order is pushed to the user terminal corresponding to the type of violation.

[0010] According to a park monitoring and management system provided by the present invention, the application operation management platform further includes: a display terminal, which is connected to the management terminal and is used to obtain and display the violation video frame and the alarm information from the management terminal, and to flash the alarm information.

[0011] According to the park monitoring and management system provided by the present invention, it further includes: a data integration platform and a data storage platform, wherein the data integration platform is connected to the management terminal and the data storage platform.

[0012] The data integration platform is used to receive the alarm information sent by the video surveillance equipment, send the alarm information to the data storage platform for storage, and send the alarm information to the management terminal.

[0013] According to a park monitoring and management system provided by the present invention, when the violation detection algorithm detects a violation in the captured video, the video surveillance device is further configured to send the algorithm information of the violation detection algorithm to the data integration platform.

[0014] The data integration platform sends the algorithm information to the data storage platform.

[0015] The data storage platform is used to associate and store the algorithm information and the corresponding alarm information.

[0016] According to the park monitoring and management system provided by the present invention, it further includes: a video transcoding platform, which is connected to the management terminal and the video monitoring equipment, and is used to transcode the illegal video frames into a video stream that conforms to the playback format of the management terminal.

[0017] According to the park monitoring and management system provided by the present invention, it further includes: an Internet of Things (IoT) platform, wherein the video transcoding platform is connected to the management terminal through the IoT platform, the video transcoding platform sends the video stream to the IoT platform, and the IoT platform is used to forward the video stream to the management terminal.

[0018] According to a park monitoring and management system provided by the present invention, the management terminal is further configured to send algorithm configuration information to the video surveillance equipment to configure the violation detection algorithm.

[0019] According to a park monitoring and management system provided by the present invention, the violation detection algorithm includes at least one of the following: personnel wearing detection algorithm, pedestrian-vehicle separation detection algorithm, personnel leaving post detection algorithm, and perimeter climbing detection algorithm.

[0020] The configuration information includes: the detection area to be configured and the pixel size of the human head image.

[0021] For the personnel absence detection algorithm, the configuration information also includes: absence mode, absence time threshold and algorithm runtime. The absence mode includes: full absence mode and minimum shift mode.

[0022] Full Absence Mode: Within the detection area, during the algorithm's runtime, if even one person leaves their post and the absence time reaches the aforementioned absence time threshold, the personnel absence detection algorithm will output absence information.

[0023] Minimum shift mode: If no one is on duty during the algorithm's runtime in the detection area, the personnel departure detection algorithm will output departure information.

[0024] For the perimeter trespass detection algorithm, the configuration information also includes detection sensitivity.

[0025] The park monitoring and management system provided by this invention deploys a violation detection algorithm on video surveillance equipment. This algorithm can automatically detect whether there are violations in the video, capture the violation video frames, generate alarm information, and send the violation video frames and alarm information to the management terminal. After receiving the violation video frames and alarm information, the management terminal generates a reminder message to promptly remind the management personnel in the monitoring room to verify and input a confirmation alarm command. After receiving the confirmation alarm command, the management terminal generates an alarm voice corresponding to the alarm information and sends the alarm voice to the corresponding alarm device for voice broadcast, thereby achieving the purpose of timely alarming and stopping violations. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of the park monitoring and management system provided by the present invention.

[0028] The attached figures are labeled as follows.

[0029] 100: Application Operation and Management Platform; 101: Management Terminal; 102: Display Terminal; 200: Video Surveillance Equipment; 300: Alarm Equipment; 400: Data Integration Platform; 500: Data Storage Platform; 600: Video Transcoding Platform; 700: Internet of Things Platform. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0031] The park monitoring and management system of this invention, such as Figure 1 As shown, it includes: video surveillance equipment 200, alarm equipment 300 and application operation management platform 100. The application operation management platform 100 includes: management terminal 101. The alarm equipment 300 and the video surveillance equipment 200 are both connected to the management terminal 101. Each video surveillance equipment 200 corresponds to one alarm equipment 300.

[0032] The video surveillance device 200 is equipped with a violation detection algorithm. When the violation detection algorithm detects a violation in the captured video, the video surveillance device 200 generates an alarm message and sends the violation video frame corresponding to the violation and the alarm message to the management terminal 101.

[0033] Specifically, the video surveillance device 200 can be a camera deployed with a video security monitoring system. The video surveillance device 200 also deploys a violation detection algorithm. The violation detection algorithm is preferably an algorithm based on an artificial intelligence detection model. Different violations correspond to different violation detection algorithms. A video surveillance device 200 can deploy multiple different violation detection algorithms. The specific violation detection algorithm deployed in the video surveillance device 200 is determined according to the area it monitors.

[0034] For example, the violation detection algorithms include: personnel wearing detection algorithm, pedestrian and vehicle separation detection algorithm, personnel leaving post detection algorithm, and perimeter climbing detection algorithm. Each area to be monitored is equipped with a video surveillance device 200. The areas to be monitored include: logistics unloading area, pedestrian and vehicle passage area on the main road in the park, fire control duty room area, and perimeter fence area of ​​the park, etc.

[0035] The video surveillance equipment 200 at the logistics unloading area is equipped with a personnel wear detection algorithm. Within the monitoring area of ​​the multi-port unloading area, if any personnel are found not wearing reflective clothing or safety helmets, the video surveillance equipment 200 will immediately generate corresponding alarm information and capture the corresponding violation video frames.

[0036] The video surveillance equipment 200 in the main road pedestrian and vehicle traffic areas within the park is equipped with a pedestrian and vehicle separation detection algorithm. Strict park safety management regulations stipulate that within the main road monitoring area, pedestrians and vehicles must travel in their designated lanes. Pedestrians must walk on the pedestrian walkways and are not allowed to cross into the vehicle lanes; similarly, vehicles must drive in the vehicle lanes and are not allowed to occupy the pedestrian walkways. If pedestrians are found walking into the vehicle lanes or vehicles are found driving in and occupying the pedestrian lanes within the main road pedestrian and vehicle traffic areas within the park, the video surveillance equipment 200 immediately generates corresponding alarm information and captures the corresponding violation video frames. For this area, a personnel wear detection algorithm can also be deployed on the corresponding video surveillance equipment 200.

[0037] A personnel absence detection algorithm is deployed on the video surveillance equipment 200 in the fire control room area. Fire control room staff are required to be on duty 24 hours a day in shifts. If any personnel leave their posts within the monitoring area of ​​the control room, resulting in an unmanned situation, the video surveillance equipment 200 will immediately generate corresponding alarm information and capture the relevant violation video frames.

[0038] The video surveillance equipment 200 around the perimeter fence of the park is equipped with a perimeter trespass detection algorithm. This is mainly to prevent dangerous individuals from climbing over the perimeter fence and entering the park, which could lead to safety accidents. Within the perimeter video monitoring area, if anyone is detected climbing or scaling the fence, the video surveillance equipment 200 immediately generates a corresponding alarm and captures the relevant violation video frames.

[0039] The management terminal 101 is used to generate a reminder message after receiving a violation video frame and alarm information, and after receiving a user's confirmation alarm command, generate an alarm voice corresponding to the alarm information, and send the alarm voice to the corresponding alarm device for voice broadcast, so as to promptly alarm and stop the violation from continuing to occur through voice broadcast.

[0040] Specifically, the management terminal 101 is equipped with an alarm information reminder module. Upon receiving a violation video frame and alarm information, the alarm information reminder module generates a reminder message to prompt the user (monitoring room staff) to verify the violation based on the violation video frame (to avoid misidentification by the violation detection algorithm). After confirmation, the user inputs a confirmation alarm command to the management terminal 101 via human-computer interaction. Upon receiving the confirmation alarm command, the management terminal 101 generates an alarm voice corresponding to the alarm information and sends the alarm voice to the corresponding alarm device 300 for voice broadcast. For example, if the alarm information is "not wearing a safety helmet," the generated alarm voice would be "Please wear your safety helmet." The management terminal 101 stores a mapping table between alarm information and corresponding alarm voices, thereby generating appropriate alarm voices based on different alarm information.

[0041] It should be noted that when deploying video surveillance equipment 200 and alarm equipment 300, the correspondence between video surveillance equipment ID and alarm equipment ID is stored in the management terminal 101 beforehand. When video surveillance equipment 200 sends a violation video frame and the alarm information to the management terminal 101, it also sends its own video surveillance equipment ID to the management terminal 101. The management terminal 101 queries the corresponding alarm equipment ID through the video surveillance equipment ID and sends the alarm voice to the corresponding alarm equipment 300 for broadcast, thereby promptly alarming against violations.

[0042] The park monitoring and management system in this embodiment deploys a violation detection algorithm on the video surveillance equipment 200. The violation detection algorithm can automatically detect whether there is any violation in the video, capture the violation video frame, generate alarm information, and send the violation video frame and alarm information to the management terminal 101. After receiving the violation video frame and alarm information, the management terminal 101 generates a reminder message to promptly remind the management personnel in the monitoring room to verify and input a confirmation alarm command. After receiving the confirmation alarm command, the management terminal 101 generates an alarm voice corresponding to the alarm information and sends the alarm voice to the corresponding alarm device for voice broadcast, thereby achieving the purpose of timely alarming and stopping violations.

[0043] In some embodiments, the video surveillance device 200 is configured to, when the violation detection algorithm detects a violation in the captured video, determine the current time as the alarm time, determine the location of the video surveillance device as the alarm location, determine the violation type (also known as: event type) based on the output of the violation detection algorithm, determine the alarm level based on the violation type or the number of violations detected by the same violation detection algorithm within a predetermined time period, and generate the alarm information based on the alarm time, alarm location, alarm level and violation type.

[0044] Specifically, when deploying video surveillance equipment 200, the location information of each video surveillance equipment 200 is set in the video surveillance equipment 200 so as to determine the alarm location when an alarm is triggered.

[0045] The type of violation can be understood as the output of the violation detection algorithm. The video surveillance equipment 200 is also used to determine the alarm level based on the violation type or the number of violations detected by the same violation detection algorithm within a predetermined time period. The alarm level can be set according to actual conditions, for example, five levels from low to high. A higher level indicates that the corresponding violation is more likely to cause a safety accident. Different violation types have different alarm levels. For example, the highest level can be set for violations such as climbing over fences, while the lowest level can be set for ordinary work positions (positions where leaving the post almost never causes a safety accident or affects production progress). The video surveillance equipment 200 is also used to count the number of violations detected by the same violation detection algorithm within a predetermined time period (which can be set according to the situation, for example, 1 hour). The more violations, the higher the alarm level. That is, for any violation type (even if its corresponding alarm level is low, repeated violations will likely cause a safety problem), the more times that violation occurs within the predetermined time period, the higher its corresponding alarm level.

[0046] In this embodiment, the alarm information includes: alarm time, alarm location, alarm level, and type of violation. This enables the staff in the monitoring room to quickly obtain relevant information about the violation. In particular, the alarm level setting for the violation allows for different handling measures to be taken according to the alarm level, such as: promptly notifying relevant management personnel to stop the violation on-site, or simply recording the alarm information.

[0047] In some embodiments, the management terminal 101 is further configured to determine whether to generate a work order based on the alarm level. If it is determined that a work order should be generated, a work order is generated based on the alarm information (i.e., the work order also includes the content of the alarm information), and the work order is pushed to the user terminal corresponding to the type of violation.

[0048] The higher the alarm level, the more serious the corresponding violation, or the more repeated and uncorrected the violation. These violations pose a significant threat to park security. Therefore, a work order is generated for alarms with high alarm levels. Specifically, the management terminal 101 is equipped with a work order configuration module, which generates work orders based on the alarm information.

[0049] Different types of violations are managed by different management departments. That is, the management terminal 101 stores the address mapping relationship between the violation type and the corresponding management department. For example, if a violation of climbing over the fence is detected, the generated work order will be pushed to the user terminal of the security department. After receiving the work order, the on-duty personnel of the security department will obtain the alarm address and stop the climbing behavior in time.

[0050] In this embodiment, by issuing work orders to the user terminals of the violation type and the corresponding management department, the relevant management personnel can be notified in a timely manner to stop the violation with a high alarm level on-site in time, so as to avoid the occurrence of safety accidents.

[0051] In some embodiments, the application operation management platform 100 further includes: a display terminal 102, which is connected to the management terminal 101 and is used to obtain and display the violation video frame and the alarm information from the management terminal 101, and to flash the alarm information.

[0052] Specifically, the display terminal 102 is a large-screen display terminal. The large-screen display terminal has a better reminder effect and can remind the on-duty personnel in the monitoring room to verify more promptly. After retrieving the video frame of the violation at that time for verification, the alarm command is confirmed in a timely manner, and the alarm is issued in conjunction with the alarm device 300. The work order is then issued, which further improves the timeliness of the alarm.

[0053] In some embodiments, the park monitoring and management system further includes a data integration platform 400 and a data storage platform 500, wherein the data integration platform 400 is connected to the management terminal 101 and the data storage platform 500.

[0054] The data integration platform 400 is used to receive the alarm information sent by the video surveillance device 200, send the alarm information to the data storage platform 500 for storage, and send the alarm information to the management terminal 101.

[0055] In this embodiment, the data integration platform 400 provides a unified data interface for each component, facilitating data interaction between the video surveillance equipment 200 and the management terminal 101. The data storage platform 500 categorizes and stores the received alarm information, making it easier for the management terminal 101 to quickly access the alarm information later.

[0056] In some embodiments, when the violation detection algorithm detects a violation in the captured video, the video surveillance device 200 is further configured to send the algorithm information of the violation detection algorithm to the data integration platform 400, the data integration platform 400 sends the algorithm information to the data storage platform 500, and the data storage platform 500 is configured to associate and store the algorithm information and the corresponding alarm information.

[0057] Specifically, algorithm information includes: algorithm name, algorithm type (e.g., image processing, natural language processing), algorithm version number, algorithm update time, and developer information. The data storage platform 500 stores the algorithm information uploaded each time a violation is detected through an algorithm information table. This means algorithm information and alarm information are uploaded synchronously, facilitating the management and traceability of the violation detection algorithm's evolution. Furthermore, the data storage platform 500 associates algorithm information with corresponding alarm information for storage. This allows for reviewing the corresponding algorithm information when alarm information is manually verified as incorrect, providing better guidance for subsequent algorithm configuration.

[0058] In some embodiments, the park monitoring and management system further includes a video transcoding platform 600, which is connected to the management terminal 101 and the video surveillance device 200, and is used to transcode the illegal video frames into a video stream that conforms to the playback format of the management terminal 101.

[0059] The video transcoding platform 600 is a comprehensive transcoding platform integrating video capture, video management, video storage, video transcoding, live video streaming, and video-on-demand. Typically, the native video format captured by the video surveillance device 200 is H.264 or H.265, which does not conform to the playback format of the management terminal 101. Therefore, in this embodiment, the video transcoding platform 600 provides the illegal video frames captured by the front-end video surveillance device 200 to itself. The video transcoding platform 600 converts the illegal video frames into FLP format and then sends them to the management terminal 101 to ensure that the management terminal 101 can play the illegal video frames.

[0060] In some embodiments, the park monitoring and management system further includes: an Internet of Things (IoT) platform 700, a video transcoding platform 600 connected to the management terminal 101 via the IoT platform 700, the video transcoding platform 600 sending the video stream to the IoT platform 700, and the IoT platform 700 forwarding the video stream to the management terminal 101.

[0061] In this embodiment, the management terminal 101 and the video transcoding platform 600 (i.e., connected to multiple video surveillance devices 200) are connected through the Internet of Things platform 700. Specifically, the Internet of Things platform 700 is used to assign an identity authentication ID to each illegal video frame, and at the same time establish an association between the video surveillance device ID of each video surveillance device 200 and the identity authentication ID of the illegal video frame sent by each device, so that the management terminal 101 can uniformly retrieve the illegal video frames corresponding to the video surveillance device ID from the Internet of Things platform 700.

[0062] In some embodiments, the management terminal 101 is further configured to send algorithm configuration information to the video surveillance device 200 to configure the violation detection algorithm. In practical applications, the violation detection algorithm needs to be configured after the video surveillance device 200 is first installed, after being disassembled and repaired and then reinstalled, or after its installation location is changed (due to changes in the visual angle and the captured image). For example, the detection area of ​​the algorithm in the captured image needs to be reconfigured. Specifically, the management terminal 101 is equipped with an algorithm configuration module. This module can receive user-entered algorithm configuration files or manually input configuration parameters, generate a configuration file containing the algorithm configuration information, and identify the video surveillance device 200 to which the algorithm configuration information is to be sent via the video surveillance device ID. It also identifies the violation detection algorithm to be configured via the algorithm name and version number, and then sends the algorithm configuration information to the corresponding violation detection algorithm in the video surveillance device 200.

[0063] In this embodiment, the management terminal 101 sends algorithm configuration information to the video surveillance device 200, which enables the video surveillance device 200 to quickly configure the violation detection algorithm after being disassembled or switched in different monitoring areas, and adapts to the algorithm configuration of the current monitoring area, making the detection of violations more accurate.

[0064] In some embodiments, the violation detection algorithm includes at least one of the following: personnel wearing detection algorithm, pedestrian-vehicle separation detection algorithm, personnel leaving post detection algorithm, and perimeter crossing detection algorithm.

[0065] The personnel detection algorithm detects individuals not wearing reflective vests or safety helmets. Through extensive sample data (images of individuals wearing reflective vests and safety helmets, and those not wearing them), and based on intelligent video analytics, it can promptly identify individuals within the monitored area who are not wearing reflective vests or safety helmets and generate alarm information. This alarm is uploaded in real time, providing evidence for tracing the source of safety incidents. For example, in the unloading area of ​​a logistics pallet, the algorithm identifies the wearing status of reflective vests and safety helmets by forklift drivers and loading / unloading operators. If they are not wearing them according to standards, an anomaly detection result is output, and the video surveillance equipment 200 generates and uploads corresponding alarm information based on this anomaly detection result.

[0066] The pedestrian-vehicle separation detection algorithm detects instances where pedestrians and vehicles deviate from their designated lanes in the main pedestrian and vehicle traffic areas within the park. Through extensive sample data (sample images of pedestrians and vehicles traveling in their respective lanes, and images of pedestrians or vehicles not traveling in their designated lanes), and based on intelligent video analysis, the algorithm can promptly detect instances of pedestrians and vehicles crossing roads within the monitored area. If a violation occurs, the algorithm outputs an anomaly detection result, and the video monitoring device 200 generates and uploads corresponding alarm information based on this anomaly detection result. For example, if a person walks into the vehicle lane, the pedestrian-vehicle separation detection algorithm outputs an anomaly detection result of "person occupying the vehicle lane," and the video monitoring device 200 generates relevant alarm information to prevent accidents involving pedestrians and vehicles.

[0067] The personnel absence detection algorithm detects whether personnel are leaving their posts in a designated work area (e.g., the fire control room area). Through simulation training with a large amount of sample data (sample images of posts with or without personnel), and based on intelligent video analysis, it can promptly detect absence behavior. If absence behavior is detected, it outputs an anomaly detection result, and the video surveillance equipment 200 generates and uploads corresponding alarm information based on this anomaly detection result.

[0068] The perimeter trespass detection algorithm detects whether anyone is trespassing over the perimeter fence of the park. Through simulation training with a large amount of sample data (images of people trespassing and images of no one trespassing), and based on intelligent video analysis, it can promptly detect whether anyone is trespassing within the monitored area. If someone is trespassing, it outputs an anomaly detection result, and the video surveillance device 200 generates and uploads corresponding alarm information based on this anomaly detection result.

[0069] The configuration information includes: the detection area to be configured and the pixel size of the human head image. The detection area is a region within the visual range of the video surveillance device 200, and the boundary contour coordinates of this region are used as the configuration input to the violation detection algorithm. For example, the visual range of the video surveillance device 200 monitoring the pedestrian and vehicle passage area on the main road in the park includes the main road and the factory buildings on both sides. The area occupied by the main road can be used as the detection area to be configured. For the human head image detection target on the detection screen, the pixel requirement is that the width and height of the head should be greater than or equal to 20 pixels, preferably greater than or equal to 40 pixels. This ensures accurate identification of personnel and makes the detected violations more accurate.

[0070] For the personnel absence detection algorithm, the configuration information also includes: absence mode, absence time threshold and algorithm runtime. The absence mode includes: full absence mode and minimum shift mode.

[0071] Full Absence Mode: Within the detection area, during the algorithm's runtime, if even one person leaves their post and the absence time reaches the aforementioned absence time threshold, the personnel absence detection algorithm will output absence information.

[0072] Minimum shift mode: If no one is on duty during the algorithm's runtime in the detection area, the personnel departure detection algorithm will output departure information.

[0073] Regardless of the off-duty mode, off-duty detection can set multiple duty areas, supporting up to 16 duty areas, and supports polygonal defense zone settings, with at least 32 sides per polygonal defense zone.

[0074] In this embodiment, by configuring different off-duty modes, the number of inspection personnel can be flexibly configured, which is suitable for off-duty inspection of different work positions. The off-duty time threshold and the algorithm runtime can be set according to different work positions. For example, for the duty position in the fire control room, the off-duty mode can be configured as the minimum duty mode, the off-duty time threshold is 20s, and the algorithm runtime is 24 hours a day, that is, it runs without stopping.

[0075] For the perimeter trespass detection algorithm, the configuration information also includes detection sensitivity. Specifically, detection sensitivity is a hyperparameter of the perimeter trespass detection algorithm, which is used to adjust the algorithm's sensitivity to identifying people trespassing over fences, so as to avoid people being identified as trespassing even if they are just standing near the fence or doing some exercise while standing on the fence.

[0076] In this embodiment, configuring different violation detection algorithms enables more accurate detection of different monitoring areas and different violations within the park.

[0077] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0078] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A park monitoring and management system, characterized in that, include: The system includes video surveillance equipment, alarm equipment, and an application operation and management platform. The application operation and management platform includes a management terminal. Both the alarm equipment and the video surveillance equipment are connected to the management terminal, and each video surveillance equipment corresponds to one alarm equipment. The video surveillance device is equipped with a violation detection algorithm. When the violation detection algorithm detects a violation in the captured video, the video surveillance device generates an alarm message and sends the violation video frame corresponding to the violation and the alarm message to the management terminal. The management terminal is used to generate a reminder message after receiving a violation video frame and alarm information, and after receiving a user's confirmation alarm command, generate an alarm voice corresponding to the alarm information and send the alarm voice to the corresponding alarm device for voice broadcast.

2. The park monitoring and management system according to claim 1, characterized in that, The video surveillance device is used to determine the current time as the alarm time and the location of the video surveillance device as the alarm location when the violation detection algorithm detects a violation in the captured video. Based on the output of the violation detection algorithm, it determines the type of violation. Based on the type of violation or the number of violations detected by the same violation detection algorithm within a predetermined time period, it determines the alarm level. Based on the alarm time, alarm location, alarm level, and violation type, it generates the alarm information.

3. The park monitoring and management system according to claim 2, characterized in that, The management terminal is also used to determine whether to generate a work order based on the alarm level. If it is determined that a work order should be generated, a work order is generated based on the alarm information, and the work order is pushed to the user terminal corresponding to the type of violation.

4. The park monitoring and management system according to claim 1, characterized in that, The application operation management platform also includes a display terminal, which is connected to the management terminal and is used to obtain and display the illegal video frame and the alarm information from the management terminal, and to flash the alarm information.

5. The park monitoring and management system according to claim 1, characterized in that, Also includes: A data integration platform and a data storage platform, wherein the data integration platform connects the management terminal and the data storage platform; The data integration platform is used to receive the alarm information sent by the video surveillance equipment, send the alarm information to the data storage platform for storage, and send the alarm information to the management terminal.

6. The park monitoring and management system according to claim 5, characterized in that, When the violation detection algorithm detects a violation in the captured video, the video surveillance device is also used to send the algorithm information of the violation detection algorithm to the data integration platform; The data integration platform sends the algorithm information to the data storage platform; The data storage platform is used to associate and store the algorithm information and the corresponding alarm information.

7. The park monitoring and management system according to claim 1, characterized in that, Also includes: A video transcoding platform, which connects the management terminal and the video monitoring equipment, is used to transcode the illegal video frames into a video stream that conforms to the playback format of the management terminal.

8. The park monitoring and management system according to claim 7, characterized in that, Also includes: The video transcoding platform is connected to the management terminal via the Internet of Things (IoT) platform. The video transcoding platform sends the video stream to the IoT platform, and the IoT platform forwards the video stream to the management terminal.

9. The park monitoring and management system according to any one of claims 1 to 8, characterized in that, The management terminal is also used to send algorithm configuration information to the video surveillance equipment to configure the violation detection algorithm.

10. The park monitoring and management system according to claim 9, characterized in that, The violation detection algorithm includes at least one of the following: personnel wearing detection algorithm, pedestrian-vehicle separation detection algorithm, personnel leaving post detection algorithm, and perimeter crossing detection algorithm; The configuration information includes: the detection area to be configured and the pixel size of the human head image; For the personnel absence detection algorithm, the configuration information also includes: absence mode, absence time threshold, and algorithm runtime period. The absence mode includes: Full Absence Mode: In the detection area, during the algorithm's runtime, if even one person leaves their post and the absence time reaches the absence time threshold, the personnel absence detection algorithm will output absence information. Minimum shift mode: If no one is on duty during the algorithm's runtime in the detection area, the personnel departure detection algorithm will output departure information. For the perimeter trespass detection algorithm, the configuration information also includes detection sensitivity.

Images