Dispatching room personnel behavior monitoring method and related device

Abstract

The invention discloses a dispatching room personnel behavior monitoring method and a related device, and the method comprises the steps: obtaining a high-definition video stream of a working region of a dispatching room in real time, and extracting a working region image; based on a preset YOLOv8 model, detection analysis is carried out according to personnel behaviors in the working area image, a behavior detection result is obtained, and the behavior detection result comprises personnel coordinates and behavior types; performing abnormal behavior analysis on the behavior detection result according to a preset behavior judgment rule to generate an abnormal event signal; and triggering an alarm mechanism according to the abnormal event signal to realize abnormal alarm, wherein the abnormal alarm comprises telephone calling and large-screen display. According to the application, the technical problem that the timeliness and intuition are difficult to ensure by personnel in a dispatching room due to poor real-time performance and low intelligent degree in the prior art can be solved.

Description

Technical Field

[0001] This application relates to the field of video surveillance technology, and in particular to a method and related device for monitoring the behavior of personnel in a dispatch room. Background Technology

[0002] Dispatch rooms, such as those for power dispatching, traffic dispatching, and emergency command and dispatching, serve as the core command hubs of various operational systems. The operational standardization and work status of dispatchers directly affect the safe, stable, and efficient operation of the entire system. Therefore, the monitoring tasks of the dispatch room are also a crucial link in ensuring the safe and stable operation of the system.

[0003] Current dispatch center monitoring solutions suffer from significant delays. Managers rely on periodic inspections or post-incident retrieval of call recordings and operation logs for checks; this approach is inefficient, cannot guarantee timeliness, and fails to ensure timely supervision and early warning. Monitoring solutions based on surveillance videos are limited to video recording and playback, requiring extensive manual review only after an incident, thus failing to provide intelligent early warning and intervention. While some dispatch systems record operation logs, they lack integration with on-site video footage, making it difficult to intuitively and comprehensively reconstruct the scene and dispatcher's state when disputes arise.

[0004] According to existing monitoring solutions, current dispatch room monitoring technology lacks real-time performance and video correlation analysis, and cannot form intelligent monitoring and early warning systems, making it difficult to grasp the scene status and personnel status in the dispatch room in a timely and intuitive manner. Summary of the Invention

[0005] This application provides a method and related device for monitoring the behavior of dispatch room personnel, which solves the technical problems of poor real-time performance and low level of intelligence in existing technologies, making it difficult to ensure the timeliness and intuitiveness of dispatch room personnel monitoring.

[0006] In view of this, the first aspect of this application provides a method for monitoring the behavior of dispatch room personnel, including:

[0007] Real-time acquisition of high-definition video streams from the dispatch room work area, and extraction of images from the work area;

[0008] Based on a preset YOLOv8 model, the behavior of people in the work area image is detected and analyzed to obtain behavior detection results, which include the coordinates of people and the type of behavior.

[0009] Based on preset behavior judgment rules, abnormal behavior analysis is performed on the behavior detection results to generate abnormal event signals;

[0010] An alarm mechanism is triggered based on the abnormal event signal to implement abnormal alarms, including telephone calls and large screen displays.

[0011] Preferably, the real-time acquisition of high-definition video streams of the dispatch room work area and the extraction of images of the work area include:

[0012] High-definition intelligent cameras are used to acquire high-frequency video streams of the dispatch room's working area in real time.

[0013] Multiple frames are extracted from the high-definition video stream based on a preset sampling frequency to generate a working area image.

[0014] Preferably, the step of detecting and analyzing human behavior in the work area image based on a preset YOLOv8 model to obtain behavior detection results includes:

[0015] The working area image is vertically cross-cut and proportionally enlarged to obtain a region slice image;

[0016] The sliced ​​image of the region is input into a preset YOLOv8 model for human behavior detection and analysis to determine the human coordinates and behavior type, and generate behavior detection results.

[0017] Preferably, the step of performing abnormal behavior analysis on the behavior detection results according to preset behavior judgment rules and generating abnormal event signals further includes:

[0018] Based on the behavior type and preset duration, alarm associations are constructed, and preset behavior judgment rules are generated.

[0019] Preferably, the step of performing abnormal behavior analysis on the behavior detection results according to preset behavior judgment rules and generating an abnormal event signal further includes:

[0020] An anomaly type label is generated based on the anomaly event signal, and associated with the corresponding segment in the high-definition video stream to obtain an anomaly-labeled video segment.

[0021] The abnormally labeled video segments are verified and stored, and alarm logs are generated.

[0022] A second aspect of this application provides a device for monitoring the behavior of dispatch room personnel, comprising:

[0023] The real-time acquisition unit is used to acquire high-definition video streams from the work area of ​​the dispatch room in real time and extract images of the work area.

[0024] The behavior detection unit is used to detect and analyze the behavior of people in the work area image based on a preset YOLOv8 model, and obtain the behavior detection result, which includes the personnel coordinates and behavior type.

[0025] An anomaly analysis unit is used to perform anomaly behavior analysis on the behavior detection results according to preset behavior judgment rules and generate anomaly event signals.

[0026] An abnormal alarm unit is used to trigger an alarm mechanism based on the abnormal event signal to realize an abnormal alarm, including telephone calls and large screen displays.

[0027] Preferably, the real-time acquisition unit is specifically used for:

[0028] High-definition intelligent cameras are used to acquire high-frequency video streams of the dispatch room's working area in real time.

[0029] Multiple frames are extracted from the high-definition video stream based on a preset sampling frequency to generate a working area image.

[0030] A third aspect of this application provides a dispatch room personnel behavior monitoring system, including: a configuration interface, a retrieval interaction interface, and any of the devices described in the second aspect;

[0031] The configuration interface is used to provide a detailed configuration interface for model parameter configuration, training parameter configuration, and decision rule configuration.

[0032] The search interface is used to provide users with a search interface for alarm logs. The search interface includes an event type input window and an event time input window.

[0033] A fourth aspect of this application provides a device for monitoring the behavior of personnel in a dispatch room, the device including a processor and a memory;

[0034] The memory is used to store program code and transmit the program code to the processor;

[0035] The processor is used to execute the dispatch room personnel behavior monitoring method described in the first aspect according to the instructions in the program code.

[0036] The fifth aspect of this application provides a computer-readable storage medium for storing program code for executing the dispatch room personnel behavior monitoring method described in the first aspect.

[0037] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0038] This application provides a method for monitoring the behavior of personnel in a dispatch room, comprising: acquiring high-definition video streams of the dispatch room's work area in real time and extracting images of the work area; performing detection and analysis of personnel behavior in the work area images based on a preset YOLOv8 model to obtain behavior detection results, the behavior detection results including personnel coordinates and behavior type; performing abnormal behavior analysis on the behavior detection results according to preset behavior judgment rules to generate abnormal event signals; and triggering an alarm mechanism based on the abnormal event signals to implement abnormal alarms, the abnormal alarms including telephone calls and large screen displays.

[0039] The dispatch room personnel behavior monitoring method provided in this application overcomes the lack of real-time performance in existing technologies by performing personnel behavior recognition and analysis on real-time acquired high-definition video streams. Furthermore, it employs the YOLOv8 model for intelligent recognition and detection of personnel behavior in the dispatch room, and then performs anomaly analysis based on preset behavior judgment rules to obtain abnormal event signals and issue alarms. This breaks through the rigid mode of passively recording video and manually reviewing it, establishing a timely and highly correlated relationship between video, personnel behavior, and alarms. This ensures the real-time, timely, and accurate nature of anomaly monitoring. Additionally, alarm methods such as telephone calls and large-screen displays allow monitoring results to be viewed more promptly and intuitively, guaranteeing good monitoring effectiveness. Therefore, this application solves the technical problems of existing technologies, such as poor real-time performance and low level of intelligence, which make it difficult to ensure the timeliness and intuitiveness of dispatch room personnel monitoring. Attached Figure Description

[0040] Figure 1 A flowchart illustrating a method for monitoring the behavior of dispatch room personnel, provided as an embodiment of this application;

[0041] Figure 2 This is a schematic diagram of the structure of a dispatch room personnel behavior monitoring device provided in an embodiment of this application. Detailed Implementation

[0042] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0043] For easier understanding, please refer to Figure 1 An embodiment of a method for monitoring the behavior of dispatch room personnel provided in this application includes:

[0044] Step 101: Acquire high-definition video streams of the dispatch room work area in real time and extract images of the work area.

[0045] Further, step 101 includes:

[0046] High-definition intelligent cameras are used to acquire high-frequency video streams of the dispatch room's working area in real time.

[0047] Multiple frames are extracted from the high-definition video stream based on a preset sampling frequency to generate a working area image.

[0048] This embodiment uses a high-definition intelligent camera to acquire real-time video of the dispatch room's working area. Multiple single-frame images are then extracted from the video to obtain the working area image. Efficient image extraction techniques can be selected during this process. The preset sampling frequency can be set according to actual conditions, such as 1 frame / 15 seconds; this is only an example and not a limitation. The image extraction operation can utilize open-source audio and video processing libraries. FFmpeg can perform video stream extraction, format conversion, and other functions. In this embodiment, this tool can be used to stably extract video frames, providing basic data for subsequent image analysis.

[0049] Step 102: Based on the preset YOLOv8 model, perform detection and analysis on the behavior of people in the work area image to obtain the behavior detection results, which include the coordinates of the people and the behavior type.

[0050] Further, step 102 includes:

[0051] The working area image is vertically cross-cut and proportionally enlarged to obtain a region slice image;

[0052] The region slice image is input into a preset YOLOv8 model for human behavior detection and analysis to determine human coordinates and behavior types, and generate behavior detection results.

[0053] To improve recognition efficiency, accuracy, and target identification, this embodiment employs dual processing at both the data and model levels. At the data level, the working area image undergoes vertical cross-cutting with a cross-cutting ratio of 10%. Assuming an image is divided into 6 parts, with height H, width W, number of parts n=6, cross-cutting ratio r, and the generated slice image index i (values ​​0, 1, 2, 3, 4, 5), then:

[0054]

[0055] in, This represents the width of each image after segmentation.

[0056] For from arrive The width of the slice image can be calculated as:

[0057]

[0058]

[0059]

[0060] in, The width of the overlapping pixels.

[0061] for The width of the slice image can be calculated as:

[0062]

[0063]

[0064] , Vertical cross-cutting does not change the image height.

[0065] Based on the calculated image width and height, the start and end coordinates of the sliced ​​image can be determined, thus achieving image segmentation of the working area. Vertical cross-segmentation of the image also avoids damage to the detection target in the image, therefore this operation not only improves target detection efficiency but also ensures detection reliability. Proportional magnification of the sliced ​​image after segmentation is then performed to make the target to be detected in the slice more prominent and clear, facilitating model analysis.

[0066] At the model level, this embodiment uses the YOLO model as the basic framework and designs a YOLOv8 model. This framework features fast detection speed and high accuracy, supports multi-target recognition and image classification, and can be applied to real-time visual analysis scenarios, adapting to the needs of this embodiment for rapid recognition of dispatcher behavior. The YOLOv8 model can accurately detect and identify people and their behaviors in region-tagged images, determining the specific location and posture of the personnel. The initial YOLOv8 model needs to be trained and adjusted using a large dataset of dispatcher behavior data before a preset YOLOv8 model can be directly used in real-time behavior detection tasks.

[0067] The dispatch room personnel behavior dataset in this embodiment includes various types of behavior data, such as sleeping, playing on mobile phones, smoking, and leaving the post. High-definition images of different types of behavior can be collected from multiple angles in the dispatch room in advance. Then, the target behavior type is labeled on these collected high-definition images on the large model platform. In addition to configuring the corresponding behavior type label, the label also needs to circle the specific posture corresponding to the behavior in the image. For example, if smoking, the fingers and cigarette are circled; if playing on a mobile phone, the hand and mobile phone are circled. After the labeling operation is completed, the behavior data training set can be generated.

[0068] The initial YOLOv8 model is trained using a behavioral data training set, enabling the model to learn how to locate personnel coordinates and recognize behaviors. To ensure the accuracy of the model, another behavioral data test set can be used to verify and analyze the trained model. Based on preset calibration standards, the optimal YOLOv8 model is selected as the preset YOLOv8 model.

[0069] It should be noted that the process of using the preset YOLOv8 model for behavior detection and analysis, in addition to obtaining the corresponding personnel coordinates and behavior type information, can also generate information such as image name and recognition time, providing more comprehensive and detailed information for subsequent anomaly analysis or retrieval. The specifics will not be elaborated here.

[0070] Step 103: Analyze the abnormal behavior of the behavior detection results according to the preset behavior judgment rules and generate abnormal event signals.

[0071] Furthermore, step 103, preceding the following, also includes:

[0072] Based on the behavior type and preset duration, alarm associations are constructed, and preset behavior judgment rules are generated.

[0073] The abnormal event signal in this embodiment includes information such as the abnormal behavior category and the abnormal behavior coordinates, which can be used as the basis for subsequent alarms and as information support for retrieval. Depending on the needs of the application platform or system, such as for ease of transmission or data processing, it may also include other necessary information, which is not specifically limited here.

[0074] The preset behavior judgment rules in this embodiment are essentially behavior anomaly analysis rules. Although the behavior types of personnel are identified through the YOLOv8 model, no direct correlation is established between the behavior types and alarm operations. Therefore, this embodiment uses preset behavior judgment rules to judge each specific behavior category, determine which behavior categories need to trigger alarms, and which alarm mechanism needs to be triggered.

[0075] The preset behavior judgment rules consist of two parts: behavior type and duration. Not all behavior types require an alarm, nor is every abnormal behavior category mandatory. For abnormal behaviors that require an alarm, they must persist for a certain period or number of occurrences before an abnormal event signal is generated and an alarm is triggered. For example, smoking can directly generate an abnormal event signal and trigger the subsequent alarm mechanism. Playing on a mobile phone can have a frequency threshold set; exceeding this threshold will be considered an abnormal event and trigger an alarm. The frequency threshold can be designed based on actual needs, such as 3 times. It's important to understand that these rules are examples; other anomaly analysis rules can be designed according to actual needs, without specific limitations.

[0076] Taking smoking, using a mobile phone, sleeping, and leaving one's post as examples, for instance, if the abnormal behavior detected is smoking (i.e., the combination of hand and cigarette features is detected, and this feature remains on the screen for more than 3 seconds), an abnormal event signal can be generated and an alarm can be triggered; if the abnormal behavior detected is using a mobile phone (i.e., the combination of hand and phone features is detected, and the distance between the phone and the person's face is less than 30cm, the duration of a single instance is more than 5 seconds, and the number of times per minute exceeds 3), an abnormal event signal can be generated; if the abnormal behavior detected is a person's head drooping more than 45 degrees... o If the eyes remain closed and the body remains still, the abnormal behavior is identified as sleeping. If this lasts for more than 10 seconds, an abnormal event signal can be generated. If no personnel are detected in the work area, it is considered an absence from duty. If this lasts for more than 15 seconds, an abnormal event signal can be generated. The above is an example of the specific abnormal event signal generation process based on preset behavior judgment rules. Specific judgment details such as time, frequency, distance, and angle can be set according to the actual situation. This is only an example and is not a limitation.

[0077] Furthermore, step 103, followed by:

[0078] Anomaly type labels are generated based on the abnormal event signals and associated with the corresponding segments in the high-definition video stream to obtain the abnormally labeled video segments.

[0079] Verify and store abnormally tagged video clips, and generate alarm logs.

[0080] Understandably, the generated abnormal event signal means that abnormal behavior has been identified and the corresponding alarm mechanism needs to be triggered. In addition, a corresponding abnormal type label can be configured for the current abnormal behavior, and then the label can be associated with the corresponding segment of the current high-definition video stream to generate an abnormal label video segment. After verification, it can be directly stored in the system to generate the corresponding alarm log, which is convenient for subsequent viewing, backtracking and handling. Specifically, it can significantly reduce the cost of manual verification and improve the efficiency of handling.

[0081] Step 104: Trigger the alarm mechanism based on the abnormal event signal to realize abnormal alarms, including telephone calls and large screen displays.

[0082] This embodiment of the anomaly alarm method includes telephone calls and large-screen displays. When the system receives an anomaly event signal, it can directly trigger the alarm mechanism, prompting on-duty personnel to handle the situation via telephone calls, and providing a direct prompt by displaying specific anomaly event information on a large screen. The prompt message may be "Anomaly XXXX occurred at seat XXXX at XXXX time. Please check the specific situation of seat XXX immediately." This information can be displayed using a red pop-up window with flashing lights. Different anomaly event display information can be generated for different anomalies, achieving adaptive and flexible alarms.

[0083] The dispatch room personnel behavior monitoring method provided in this application overcomes the lack of real-time performance in existing technologies by performing personnel behavior recognition and analysis on real-time acquired high-definition video streams. Furthermore, it employs a YOLOv8 model for intelligent recognition and detection of personnel behavior in the dispatch room, and then performs anomaly analysis based on preset behavior judgment rules to obtain abnormal event signals and issue alarms. This breaks through the current rigid mode of passively recording videos and manually reviewing them, establishing a timely and highly correlated relationship between video, personnel behavior, and alarms. It ensures the real-time, timely, and accurate nature of anomaly monitoring. Additionally, alarm methods such as telephone calls and large-screen displays allow monitoring results to be viewed more promptly and intuitively, ensuring good monitoring effectiveness. Therefore, this application embodiment can solve the technical problems of poor real-time performance and low intelligence level in existing technologies, which makes it difficult to ensure the timeliness and intuitiveness of dispatch room personnel monitoring.

[0084] For easier understanding, please refer to Figure 2 This application provides an embodiment of a dispatch room personnel behavior monitoring device, comprising:

[0085] The real-time acquisition unit 201 is used to acquire high-definition video streams of the dispatch room work area in real time and extract images of the work area.

[0086] The behavior detection unit 202 is used to detect and analyze the behavior of people in the work area image based on a preset YOLOv8 model, and obtain the behavior detection results, which include the personnel coordinates and behavior type.

[0087] The anomaly analysis unit 203 is used to perform abnormal behavior analysis on the behavior detection results according to the preset behavior judgment rules and generate abnormal event signals.

[0088] The abnormal alarm unit 204 is used to trigger the alarm mechanism based on abnormal event signals to realize abnormal alarms, including telephone calls and large screen displays.

[0089] Furthermore, the real-time acquisition unit 201 is specifically used for:

[0090] High-definition intelligent cameras are used to acquire high-frequency video streams of the dispatch room's working area in real time.

[0091] Multiple frames are extracted from the high-definition video stream based on a preset sampling frequency to generate a working area image.

[0092] This application also provides a dispatch room personnel behavior monitoring system, including: a configuration interface, a retrieval interaction interface, and any of the devices described in the above device embodiments;

[0093] The configuration interface provides detailed configuration options for model parameter configuration, training parameter configuration, and decision rule configuration.

[0094] The search interface provides users with a search interface for alarm logs. The search interface includes an event type input window and an event time input window.

[0095] This embodiment constructs a complete monitoring and alarm system for abnormal personnel behavior in the dispatch room through a closed-loop architecture of data acquisition, intelligent identification and analysis, application alarms, and backend management, combined with deep learning models and multi-platform linkage mechanisms. Specifically, the real-time acquisition unit can acquire high-definition video streams of the dispatch room's work area in real time and extract images of the work area; the behavior detection unit can detect and analyze personnel behavior in the work area images based on a preset YOLOv8 model, obtaining behavior detection results including personnel coordinates and behavior type; the anomaly analysis unit can perform abnormal behavior analysis on the behavior detection results according to preset behavior judgment rules, generating abnormal event signals; and the anomaly alarm unit can trigger an alarm mechanism based on the abnormal event signals to implement anomaly alarms, including telephone calls and large-screen displays.

[0096] The configuration interface and search interface enhance the system's interactivity. Specifically, the configuration interface can be adapted to different dispatch rooms, such as those for power, transportation, and emergency command, to meet their individual needs; while the search interface improves the efficiency and standardization of retrospective viewing.

[0097] This embodiment overcomes the information silo problem of existing technologies. Current monitoring systems, telephone systems, and large-screen display systems operate independently and cannot effectively coordinate. Even when anomalies are detected, alarm methods are limited, failing to ensure timely and effective delivery of information to responsible parties. This embodiment can promptly detect and analyze abnormal behavior in real-time video, determine and analyze abnormal behavior, and trigger alarms using different mechanisms, ensuring the timeliness and intuitiveness of monitoring the behavior of personnel in the dispatch room.

[0098] This application also provides a dispatch room personnel behavior monitoring device, the device including a processor and a memory;

[0099] The memory is used to store program code and transfer the program code to the processor;

[0100] The processor is used to execute the dispatch room personnel behavior monitoring method in the above method embodiment according to the instructions in the program code.

[0101] This application also provides a computer-readable storage medium for storing program code for executing the dispatch room personnel behavior monitoring method in the above method embodiments.

[0102] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0103] 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0104] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0105] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of this application through a computer device (which may be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0106] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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. Such 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 this application.

Claims

1. A method for monitoring the behavior of dispatch room personnel, characterized in that, include: Real-time acquisition of high-definition video streams from the dispatch room work area, and extraction of images from the work area; Based on a preset YOLOv8 model, the behavior of people in the work area image is detected and analyzed to obtain behavior detection results, which include the coordinates of people and the type of behavior. Based on preset behavior judgment rules, abnormal behavior analysis is performed on the behavior detection results to generate abnormal event signals; An alarm mechanism is triggered based on the abnormal event signal to implement abnormal alarms, including telephone calls and large screen displays.

2. The method for monitoring the behavior of dispatch room personnel according to claim 1, characterized in that, The real-time acquisition of high-definition video streams of the dispatch room's working area and the extraction of images from the working area include: High-definition intelligent cameras are used to acquire high-frequency video streams of the dispatch room's working area in real time. Multiple frames are extracted from the high-definition video stream based on a preset sampling frequency to generate a working area image.

3. The method for monitoring the behavior of dispatch room personnel according to claim 1, characterized in that, The method of detecting and analyzing human behavior in the work area image based on a preset YOLOv8 model to obtain behavior detection results includes: The working area image is vertically cross-cut and proportionally enlarged to obtain a region slice image; The sliced ​​image of the region is input into a preset YOLOv8 model for human behavior detection and analysis to determine the human coordinates and behavior type, and generate behavior detection results.

4. The method for monitoring the behavior of dispatch room personnel according to claim 1, characterized in that, The step of performing abnormal behavior analysis on the behavior detection results according to preset behavior judgment rules and generating abnormal event signals, before which the following is also included: Based on the behavior type and preset duration, alarm associations are constructed, and preset behavior judgment rules are generated.

5. The method for monitoring the behavior of dispatch room personnel according to claim 1, characterized in that, The step of performing abnormal behavior analysis on the behavior detection results according to preset behavior judgment rules and generating an abnormal event signal, further includes: An anomaly type label is generated based on the anomaly event signal, and associated with the corresponding segment in the high-definition video stream to obtain an anomaly-labeled video segment. The abnormally labeled video segments are verified and stored, and alarm logs are generated.

6. A device for monitoring the behavior of personnel in a dispatch room, characterized in that, include: The real-time acquisition unit is used to acquire high-definition video streams from the work area of ​​the dispatch room in real time and extract images of the work area. The behavior detection unit is used to detect and analyze the behavior of people in the work area image based on a preset YOLOv8 model, and obtain the behavior detection result, which includes the personnel coordinates and behavior type. An anomaly analysis unit is used to perform anomaly behavior analysis on the behavior detection results according to preset behavior judgment rules and generate anomaly event signals. An abnormal alarm unit is used to trigger an alarm mechanism based on the abnormal event signal to realize an abnormal alarm, including telephone calls and large screen displays.

7. The dispatch room personnel behavior monitoring device according to claim 6, characterized in that, The real-time acquisition unit is specifically used for: High-definition intelligent cameras are used to acquire high-frequency video streams of the dispatch room's working area in real time. Multiple frames are extracted from the high-definition video stream based on a preset sampling frequency to generate a working area image.

8. A system for monitoring the behavior of personnel in a dispatch room, characterized in that, include: The configuration interface, the search interaction interface, and the apparatus described in any of claims 6-7; The configuration interface is used to provide a detailed configuration interface for model parameter configuration, training parameter configuration, and decision rule configuration. The search interface is used to provide users with a search interface for alarm logs. The search interface includes an event type input window and an event time input window.

9. A device for monitoring the behavior of personnel in a dispatch room, characterized in that, The device includes a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the dispatch room personnel behavior monitoring method according to any one of claims 1-5 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code for executing the dispatch room personnel behavior monitoring method according to any one of claims 1-5.

Images