Detection Method, System, Device and Medium for Detecting Intruding Persons in Rail Transit Supervision Area

Abstract

The present application relates to the technical field of rail transit safety monitoring, and discloses a detection method, system, device and medium for intruders in a rail supervision area, including: performing intrusion detection on a first supervision area and a second supervision area of a rail transit station, wherein the first supervision area includes an area around the train track where train attendants are allowed to conduct on-site inspections when the train arrives at the station, and the second supervision area includes a prohibited stay area around the train track except for the first supervision area; in response to the presence of an arriving train at the rail transit station and the target door of the arriving train being in the door open state, the intrusion detection of the first supervision area is turned off. The present application identifies the open / closed state of the target door to determine whether to turn off or turn on the monitoring of the first supervision area, avoiding false alarms caused by misjudging personnel intrusion when train attendants get off the train for on-site inspections, realizing flexible zonal monitoring, which can not only avoid false alarms and reduce the false alarm rate, but also effectively prevent missed detections.

Description

Technical Field

[0001] This application relates to the technical field of rail transit safety monitoring, and particularly to a detection method, system, device and medium for detecting intruders in a rail supervision area. Background Art

[0002] With the rapid development of urban rail transit, subways, high-speed rails, light rails, etc. have been widely used in urban travel as convenient public transportation tools. However, the safety problems of these rail transit tools have become increasingly prominent, especially in the detection of illegal intrusion of personnel in the supervision area. For example, when the subway arrives at the station, the behavior of the conductor or train attendant getting off the train to patrol increases the complexity of safety detection. If only relying on a single detection method in the existing technology, it may misjudge the conductor or train attendant as a dangerous intruder, resulting in unnecessary safety alarms and patrol interference. Summary of the Invention

[0003] The main purpose of this application is to provide a detection method, system, device and medium for detecting intruders in a rail supervision area, aiming to solve the technical problem of misjudging the train attendant as an intruder in the existing technology, resulting in false alarms.

[0004] In the first aspect of this application, a detection method for detecting intruders in a rail transit supervision area is provided. The detection method for detecting intruders in a rail transit supervision area includes:

[0005] Performing intrusion detection on the first supervision area and the second supervision area of a rail transit station. Among them, the first supervision area includes the area around the train track where the train attendant is allowed to patrol when the train arrives at the station, and the second supervision area includes the prohibited stay area around the train track except the first supervision area;

[0006] In response to the presence of an arriving train at the rail transit station and the target door of the arriving train being in the door open state, the intrusion detection of the first supervision area is closed.

[0007] This application also provides a detection system for detecting intruders in a rail transit supervision area. The detection system for detecting intruders in a rail transit supervision area includes:

[0008] A detection control module for performing intrusion detection on the first supervision area and the second supervision area of a rail transit station. Among them, the first supervision area includes the area around the train track where the train attendant is allowed to patrol when the train arrives at the station, and the second supervision area includes the prohibited stay area around the train track except the first supervision area;

[0009] The detection control module is further configured to close the intrusion detection of the first supervision area in response to the presence of an arriving train at the rail transit station and the target door of the arriving train being in the door open state.

[0010] A third aspect of the present application provides a computer device, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor calls the instructions in the memory to enable the computer device to execute the above-mentioned method for detecting intruders in the rail transit supervision area.

[0011] A fourth aspect of the present application provides a computer-readable storage medium, in which instructions are stored. When the instructions are run on a computer, the computer is enabled to execute the above-mentioned method for detecting intruders in the rail transit supervision area.

[0012] In this embodiment, by identifying and monitoring the opening and closing state of the target door, it is determined whether to temporarily close the monitoring of the first supervision area, thus avoiding misjudging personnel intrusion and resulting in false alarms when the conductor gets off the train for inspection; during this period, the monitoring of the second supervision area is always maintained, achieving flexible partition monitoring. This not only avoids false alarms as much as possible and reduces the false alarm rate, but also effectively prevents missed detections, improving the intelligent level of rail transit safety monitoring and management; this embodiment can accurately distinguish the staying behaviors of staff and non-staff under various conditions of different lighting conditions and changes in the position of the train head, ensuring that no false alarms occur when the staff is patrolling, avoiding unnecessary interference, and providing reliable guarantee for rail transit operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic flowchart of the first embodiment of the method for detecting intruders in the rail transit supervision area in the embodiment of the present application;

[0014] Figure 2 It is a schematic diagram of the positions of the track, the first supervision area and the second supervision area in the embodiment of the present application;

[0015] Figure 3 It is a scene diagram of the track, the first supervision area and the second supervision area in the embodiment of the present application;

[0016] Figure 4 It is a schematic diagram of the normal patrol state of the conductor in the embodiment of the present application;

[0017] Figure 5 It is a schematic diagram of simulating an intruder entering the second supervision area in the embodiment of the present application;

[0018] Figure 6 It is a schematic flowchart of the second embodiment of the method for detecting intruders in the rail transit supervision area in the embodiment of the present application;

[0019] Figure 7 It is a schematic diagram of the functional modules of an embodiment of the detection system for intruders in the rail transit supervision area in the embodiment of the present application;

[0020] Figure 8Schematic diagram of an embodiment of a computer device in an embodiment of the present application. Detailed implementation manners

[0021] Terms such as "first", "second", "third", "fourth", etc. (if any) in the description, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily limit to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0022] With the rapid development of urban rail transit, rail transit such as subways, as a convenient public transportation tool, has been widely used in urban travel. However, the safety issues in the front area of the subway train are becoming increasingly prominent, especially in the timely detection of personnel climbing over the platform screen doors. When the subway arrives at the station, the behavior of the train crew getting off the train to patrol increases the complexity of safety detection. If only relying on a single detection method, it may misjudge the train crew as a dangerous intruder, resulting in unnecessary safety alarms and patrol disruptions.

[0023] Based on this, the present application provides a detection solution for intruders in the rail transit supervision area.

[0024] Reference Figure 1 , in an embodiment, the present application provides a detection method for intruders in the rail transit supervision area. The detection method for intruders in the rail transit supervision area includes:

[0025] S100: Perform intrusion detection on the first supervision area and the second supervision area of the rail transit station. Among them, the first supervision area includes the area around the train track where the train crew is allowed to get off the train for inspection, and the second supervision area includes the prohibited stay area around the train track except the first supervision area.

[0026] Specifically, the rail transit in this embodiment includes public transportation such as subways, high-speed rails, light rails, trains, and bullet trains, which are not limited to this.

[0027] The supervision area is the key supervision area of the rail transit station. For example, in the front area, there are the first supervision area and the second supervision area. The first supervision area is the area where the train crew (or, the train driver) is allowed to get off the train to patrol, inspect, and command, and it is a prohibited area for illegal personnel such as passengers to stay. The second supervision area is the area where any personnel, especially passengers, are usually prohibited from staying.

[0028] Both the first supervision area and the second supervision area are areas adjacent to the track and are used by the station staff. Under normal circumstances, illegal personnel such as passengers are not allowed to stay in the first supervision area and the second supervision area.

[0029] When there is no train arriving and staying on the track of the rail transit station, both the first supervision area and the second supervision area need to be monitored. Therefore, it is necessary to simultaneously enable the intrusion detection function for the first supervision area and the second supervision area of the rail transit station.

[0030] Figure 2 It is a schematic diagram of the positions of the track, the first supervision area, and the second supervision area in the embodiment of the present application; refer to Figure 2 , the first supervision area and the second supervision area are areas beside the track. Under normal circumstances, the train attendant can stand in the first supervision area when getting off the train for patrol.

[0031] Among them, the intrusion detection function for the first supervision area and the second supervision area of the rail transit station can be realized through computer vision technology.

[0032] If, when there is no train arriving, it is detected that there is a person intrusion in at least one of the first supervision area and the second supervision area, an alarm will be given.

[0033] Taking the rail transit as the subway as an example, the first supervision area and the second supervision area can be the area between the track at the head of the train and the platform screen door. If someone climbs over the platform screen door and enters this area, there may be danger. Therefore, it is necessary to monitor whether there is an illegal person intrusion in this area.

[0034] S200: In response to the presence of an arriving train at the rail transit station and the target door of the arriving train being in the door-open state, the intrusion detection for the first supervision area is turned off.

[0035] Specifically, the target door is the carriage door for the train attendant or the conductor to get on and off the train. If it is detected that there is an arriving train at the rail transit station and the target door of this arriving train is in the door-open state, it means that the train attendant or the conductor may get off the train for inspection, patrol, or command, etc., and the train attendant or the conductor may stay in the first supervision area. Therefore, in order to prevent misjudgment and false alarm, in this case, the intrusion detection function for the first supervision area is turned off, and at the same time, the intrusion detection function for the second supervision area is maintained.

[0036] In this embodiment, by identifying and monitoring the opening and closing states of the target door, it is determined whether to temporarily close the monitoring of the first supervision area, thereby avoiding misjudging personnel intrusion when the train attendant gets off the train for inspection, resulting in false alarms. During this period, the monitoring of the second supervision area is always maintained, achieving flexible zonal monitoring. This not only avoids false alarms as much as possible, reduces the false alarm rate, but also effectively prevents missed detections, improving the intelligent level of rail transit safety monitoring and management. This embodiment can accurately distinguish the lingering behaviors of staff and non-staff under various conditions of different lighting conditions and changes in the position of the train head, ensuring that no false alarms occur when the staff is patrolling, avoiding unnecessary interference, and providing reliable guarantee for rail transit operation.

[0037] In one embodiment, after closing the intrusion detection of the first supervision area in step S200 in response to the arrival of a train at a rail transit station and the target door of the arriving train being in the door-open state, the method for detecting an intruder in the rail transit supervision area further includes:

[0038] In response to the target door changing from the door-open state to the door-closed state, the intrusion detection of the first supervision area is restarted.

[0039] Specifically, if it is detected that the target door is in the door-closed state, it means that the train attendant or the conductor has boarded the train. Under normal circumstances, the train attendant or the conductor will not continue to linger in the first supervision area. Therefore, it is necessary to restart the intrusion detection function of the first supervision area. In this case, the intrusion detection functions of both the first supervision area and the second supervision area are enabled.

[0040] After enabling the intrusion detection function of the first supervision area, if a personnel intrusion is detected in the first supervision area again, it means that there is an illegal intrusion by a person other than the train attendant or the conductor in the first supervision area, and an alarm needs to be issued, and no misjudgment will occur.

[0041] In this embodiment, by identifying and monitoring the opening and closing states of the target door, it is determined whether to temporarily close the monitoring of the first supervision area, thereby avoiding misjudging personnel intrusion when the train attendant gets off the train for inspection, resulting in false alarms. In this embodiment, the monitoring of the first supervision area can be restarted in time after the target door is closed again. During this period, the monitoring of the second supervision area is always maintained, achieving flexible zonal monitoring. This not only avoids false alarms as much as possible, reduces the false alarm rate, but also effectively prevents missed detections, improving the intelligent level of rail transit safety monitoring and management. This embodiment can accurately distinguish the lingering behaviors of staff and non-staff under various conditions of different lighting conditions and changes in the position of the train head, ensuring that no false alarms occur when the staff is patrolling, avoiding unnecessary interference, and providing reliable guarantee for rail transit operation.

[0042] In one embodiment, the method for detecting an intruder in the rail transit supervision area further includes:

[0043] In response to the target vehicle door being in the open state and there being a personnel intrusion in the second supervision area, a first warning is given.

[0044] And / or,

[0045] In response to there being an intruder in at least one of the first supervision area and the second supervision area, a second warning is given.

[0046] Specifically, when the target vehicle door is in the open state, if it is detected that there is a personnel intrusion in the second supervision area, a first warning will be given.

[0047] Computer vision technology can be used to detect intrusion in the second supervision area. More specifically, a camera is used to take a picture of the area including the supervision area to obtain a captured picture, or, the area including the supervision area is videoed to obtain a video.

[0048] The target detection model is used to perform target detection and classification on the captured picture, locate the second supervision area in the captured picture, and determine whether there is a personnel intrusion or loitering in the second supervision area.

[0049] Or,

[0050] Video frames (frame images) are intercepted from the video, and the target detection model is used to perform target detection and classification on the video frames, locate the second supervision area in the video frames, and determine whether there is a personnel intrusion or loitering in the second supervision area.

[0051] Among them, intercepting video frames from the video can be frame-by-frame interception or interval interception at a preset time interval. This application does not limit this. Among them, interval interception at a preset time interval can ensure the diversity of pictures, reduce duplicate pictures, and avoid pictures being too repetitive and dense.

[0052] If it is detected that there is a personnel intrusion in the second supervision area, a first warning is given.

[0053] If it is not detected that there is a personnel intrusion in the second supervision area, there is no need to give a first warning.

[0054] Among them, the first warning may include but is not limited to at least one of the following: activating the alarm system of the rail transit station to issue an alarm to instruct the staff to handle the personnel intrusion in a timely manner and clear the intruders; displaying the target picture on the monitoring system, boxing the intruders and / or flashing red and orange alternately on the target picture for warning, where the target picture is the captured picture or video frame in which it is detected that there is a personnel intrusion in the second supervision area; showing the video to the staff and flashing red and orange alternately in the video for warning, etc.

[0055] After the target door is in the door closed state, the monitoring system will perform intrusion detection on the first supervision area and the second supervision area. If there is a person intrusion in at least one of these two supervision areas, a second warning will be issued.

[0056] At this time, the second warning will not misjudge the train conductor or the conductor as an intruder. Therefore, misjudgment is avoided.

[0057] Among them, computer vision technology can be used to perform intrusion detection on the first supervision area and the second supervision area. More specifically, a camera is used to take a picture of the area including the supervision area to obtain a captured picture, or, the area including the supervision area is filmed to obtain a filmed video.

[0058] The target detection model is used to perform target detection and classification on the captured picture, locate the first supervision area and the second supervision area in the captured picture, and determine whether there is a person intrusion or stay in the first supervision area and the second supervision area.

[0059] Or,

[0060] Video frames (frame images) are intercepted from the filmed video, and the target detection model is used to perform target detection and classification on the video frames, locate the first supervision area and the second supervision area in the video frames, and determine whether there is a person intrusion or stay in the first supervision area and the second supervision area.

[0061] Among them, intercepting video frames from the filmed video can be frame-by-frame interception or interval interception at a preset time interval. This application does not limit this.

[0062] If it is detected that there is a person intrusion in the first supervision area and / or the second supervision area, a second warning will be issued.

[0063] If it is not detected that there is a person intrusion in the first supervision area and the second supervision area, there is no need to issue a second warning.

[0064] Among them, the second warning may include but is not limited to at least one of the following: activating the alarm system of the rail transit station to issue an alarm to instruct the staff to handle the person intrusion in time and clear the intruder; displaying the target picture on the monitoring system, boxing the intruder and / or flashing red and orange alternately on the target picture to indicate a warning, where the target picture is the captured picture or video frame when it is detected that there is a person intrusion in the first supervision area or the second supervision area; displaying the filmed video to the staff and flashing red and orange alternately in the filmed video to indicate a warning, etc.

[0065] In this embodiment, by identifying and monitoring the opening and closing state of the target door, it is determined whether to temporarily close the monitoring of the first supervision area, thus avoiding misjudgment of personnel intrusion when the conductor gets off the train for inspection, resulting in false alarms; during this period, the monitoring of the second supervision area is always maintained, achieving flexible zonal monitoring, which not only avoids false alarms as much as possible, reduces the false alarm rate, but also effectively prevents missed inspections, improving the intelligent level of rail transit safety monitoring and management; this embodiment can accurately distinguish the staying behaviors of staff and non-staff under various conditions of different lighting conditions and changes in the position of the train head, ensuring that no false alarms occur when the staff patrols, avoiding unnecessary interference, and providing reliable guarantee for rail transit operation. In different supervision periods, this embodiment can give timely warnings for the second supervision area with personnel intrusion, or, give timely warnings for both the first supervision area and the second supervision area with personnel intrusion.

[0066] Taking rail transit as an example of subway, it intelligently detects whether there is any behavior of non-conductor personnel or non-staff climbing over the platform screen door in the subway train head area. To achieve this goal, it can be judged by monitoring whether the human detection center point is located in the set intrusion area. To effectively narrow the area where the conductor gets off the train for patrol, the intrusion area is divided into two parts: Area 1 and Area 2. Among them, Area 1 is the area where the conductor is allowed to enter. The system is designed such that only when the subway arrives at the station and the door is open, the conductor is allowed to enter Area 1 without triggering an alarm, which can shorten the detection time of closing Area 1. The specific detection opening and closing logic of Area 1 and Area 2 is described in the above method. Through this division, the system can more accurately monitor the intrusion behavior of non-conductors, thus ensuring that no false alarms occur when the conductor patrols normally and avoiding unnecessary interference. In addition, the subway arrival detection and the train head door opening and closing detection can be simplified to a comprehensive judgment of the door state of the train head. Through this solution, the safety monitoring ability of the subway train head area can be significantly improved, the false alarm rate can be reduced, and the stability and safety of operation can be ensured.

[0067] Figure 3 It is a scene diagram of the track, the first supervision area and the second supervision area in the embodiment of the present application; Figure 4 It is a schematic diagram of the normal patrol state of the conductor in the embodiment of the present application; Figure 5 It is a schematic diagram of simulating an intruder entering the second supervision area in the embodiment of the present application; Through Figures 3 - 5 It can be seen that under normal circumstances, when the train head door is open, the conductor or the train captain can stay temporarily in the first supervision area, and at this time, the monitoring of the first supervision area is closed. In the case of no train arrival at the station, both the first supervision area and the second supervision area should be monitored, and there should be no personnel staying in both the second supervision area and the first supervision area.

[0068] The detection method for intruders in the rail transit supervision area of this embodiment can be applied to, for example, the detection of personnel climbing over the platform screen door at the front of the subway train. It can accurately distinguish the intrusion behaviors of train conductors or drivers from those of non-staff members under different lighting conditions and changes in the position of the train head, thereby improving the intelligent level of subway safety management and providing more reliable guarantees for operation.

[0069] The detection method for intruders in the rail transit supervision area of this application can be applied to urban rail transit systems, especially in the field of safety monitoring of subway operations. Its technology can be widely used in intrusion detection in the front area of the subway train to ensure that unauthorized personnel cannot easily climb over the platform screen door and guarantee the safety of passengers and staff. In addition, this method is also applicable to the safety management of other public transportation tools, such as light rail and high-speed rail, as well as the monitoring and management of large public places, with a wide range of applications.

[0070] In one embodiment, the detection method for intruders in the rail transit supervision area further includes: if it is detected that a train arrives at the station, monitoring the door opening and closing state of the target door of the arriving train, where the door opening and closing state is the door open state or the door closed state.

[0071] Specifically, if the arriving train includes a driver's door and passenger doors, the target door is the driver's door.

[0072] If all the door types of the arriving train are the same, or the train conductor (driver) and passengers can be in the same carriage, the train conductor or driver may get off the train from any door, then the target door can be any door.

[0073] When a train arrives at the station, the arriving train can automatically open the carriage doors. Therefore, the arriving train can send the door opening and closing state to the monitoring system in a communication manner when the doors are opened or after the doors are opened.

[0074] Or,

[0075] The monitoring system can monitor the opening and closing state of the target door of the arriving train through computer vision technology.

[0076] In one embodiment, before closing the intrusion detection of the first supervision area, the detection method for intruders in the rail transit supervision area further includes:

[0077] Inputting the video frames intercepted from the camera video data into a trained target detection model, and performing target door detection through the trained target detection model;

[0078] Determining the door opening and closing state of the target door according to the obtained target detection result, where the door opening and closing state includes the door open state and the door closed state.

[0079] Specifically, the detection method for intruders in the rail transit supervision area in this embodiment is an intelligent detection method for intruders in the rail transit supervision area based on computer vision.

[0080] Both the first supervision area and the second supervision area are beside the train tracks of the rail transit station. Therefore, cameras can be deployed near the tracks. If a train arrives at the station, the camera can capture a first captured picture or a first captured video that includes the target door of the arriving train. Additionally, the first captured video data can also be the captured video after determining that the train has arrived at the station.

[0081] Of course, in a specific embodiment, the shooting angle of the camera can also be set so that the camera can capture the train tracks (which can capture the target door of the arriving train), the first supervision area, and the second supervision area in the same video. That is, the first captured video can also include the first supervision area and the second supervision area, etc. This application does not limit this. This can reduce the use and waste of cameras while ensuring the time consistency of the detection of the arriving train, the target door, and the supervision area.

[0082] To prevent missed shots, this embodiment preferably captures a captured video and transmits the captured video as the captured video data to the computer device of the monitoring system.

[0083] The computer device can intercept video frames from the captured video data frame by frame or at a predetermined time interval. This embodiment does not limit this.

[0084] The video frames intercepted from the captured video data are sequentially input into the trained target detection model in the order of shooting. The trained target detection model can identify and detect the target doors in different switch states of the train. For example, it can identify the target doors in the door - open state and also the target doors in the door - closed state, and both are the targets to be detected.

[0085] In addition, if the target door is the driver's door, which is different from the passenger door, the target detection model also has the function of identifying and distinguishing the target door from the passenger door.

[0086] The trained object detection model performs object detection or object capture on the target door in the door open state and the target door in the door closed state in the video frame, frames the target area or the candidate area containing the target, and classifies the target area or the candidate area to obtain the object detection result. Among them, the object detection result includes the object type and the calibration coordinates of the detected object. The object detection result indicates the object type of the detected object in the video frame, and the object type is the target door in the door open state or the target door in the door closed state. Therefore, according to the object detection result, the door switch state of the target door can be determined.

[0087] In a specific embodiment, the object detection model can be constructed based on, for example, the detection algorithm of the YOLO (You Only Look Once) series. The detection algorithm is, for example, one of YOLO V1, YOLO V5, YOLO V7, YOLO V8, etc.

[0088] Taking the subway as an example, when the subway arrives at the station, the system will analyze the video captured by the on-site camera frame by frame to detect the state of the long door of the train. By analyzing the detection results, the system can accurately judge the arrival situation of the subway and the switch state of the door. When the long door of the train is detected, it indicates that the subway has arrived at the station; otherwise, it means that the subway has not arrived yet. After the subway arrives at the station, the system will continuously monitor the opening state of the door. If the long door of the train is opened, the system will allow the station staff to patrol and suspend the intrusion detection function for Area 1 (the first supervision area) according to the process design. Subsequently, the system will continuously monitor the state of the long door of the train until the door is closed again, at which time the intrusion detection of Area 1 will be restarted.

[0089] In this embodiment, the object detection model can accurately perform object detection on the target door in different switch states, improving the efficiency and reliability of the overall safety management.

[0090] In one embodiment, in step S200, in response to the presence of an arriving train at the rail transit station and the target door of the arriving train being in the door open state, closing the intrusion detection for the first supervision area includes: in response to the presence of an arriving train at the rail transit station and the target door being monitored to be in the door open state in a continuous first number of first video frames, closing the intrusion detection for the first supervision area.

[0091] Specifically, the captured video contains multiple video frames. To prevent misdetection of the door opening and closing state of the target door, several video frames can be sequentially input into the target detection model according to the chronological order of the timestamps of the video frames. If after continuously inputting the first quantity of first video frames into the target detection model, the target detection results of these first video frames all indicate that the target door is in the door open state, then the intrusion detection function for the first supervision area is determined to be closed. This not only ensures the accurate determination of the door opening and closing state of the target door and prevents misjudgment, but also prevents the misclosing of the intrusion detection function for the first supervision area, resulting in missed detections.

[0092] Through the above measures, this application can ensure the precise monitoring of the state of the subway car long door, improving the efficiency and reliability of the overall safety management.

[0093] In this embodiment, through continuous detection, when the preset quantity is reached, it is confirmed whether the target door is open or the target door is closed, effectively reducing the false alarm rate.

[0094] In one embodiment, before closing the intrusion detection for the first supervision area, the method for detecting intrusion of personnel in the rail transit supervision area further includes:

[0095] Input the video frames intercepted from the captured video data into the trained target detection model, and perform target door detection through the trained target detection model;

[0096] Determine the door opening and closing state of the target door according to the obtained target detection result;

[0097] In response to the target door changing from the door open state to the door closed state, the intrusion detection for the first supervision area is restarted, including: if it is monitored that the target door is in the door closed state in consecutive second video frames, then the intrusion detection function for the first supervision area is restarted, where the time points of the second video frames are later than those of the first video frames.

[0098] Specifically, the video frames are input into the target detection model according to the chronological order of the timestamps. Therefore, under normal circumstances, the target door will change from the door closed state (such as when just entering the station) to the door open state (midway) and finally to the door closed state (such as when about to depart from the station).

[0099] Input several video frames into the target detection model in sequence. If, after continuously inputting the second quantity of second video frames into the first target detection model, the target detection results of these second video frames all indicate that the target door is in the closed state, and it is switched from the open state to the closed state, then determine to restart the intrusion detection function for the first supervision area. This not only ensures the accurate determination of the door opening and closing state of the target door, preventing misjudgment, but also prevents the intrusion detection function for the first supervision area from being accidentally started when the conductor or the train attendant gets off the train for inspection or before boarding, resulting in false alarms.

[0100] Among them, the above first quantity and second quantity are specifically configured according to the actual application scenario, and this embodiment does not limit this.

[0101] Through the above measures, this application can ensure the precise monitoring of the door state of the subway car door, improving the efficiency and reliability of the overall safety management.

[0102] In this embodiment, through continuous detection, when the preset quantity is reached, it is confirmed whether the target door is open or the target door is closed, effectively reducing the false alarm rate.

[0103] In one embodiment, the method for detecting intruders in the rail transit supervision area further includes: monitoring whether there is an arriving train at the rail transit station.

[0104] Specifically, the method for detecting intruders in the rail transit supervision area of this embodiment can be applied to the monitoring system of the rail transit station. It can determine whether a train has arrived by communicating with the arriving train. More specifically, the arriving train will send an arrival message to the monitoring system of the rail transit station when entering the station or stopping at the station, so that the monitoring system can timely determine that the train has arrived.

[0105] Of course, the monitoring system can also monitor whether there is an arriving train or an incoming train on the track through computer vision technology, so as to determine whether there is an arriving train on the track of the rail transit station.

[0106] It should be noted that in different application scenarios, the arriving train may be a subway, a high-speed train, a bullet train, a light rail, etc. This application does not limit this.

[0107] In one embodiment, before closing the intrusion detection for the first supervision area, the method for detecting intruders in the rail transit supervision area further includes:

[0108] Input the video frames intercepted from the camera video data into the trained target detection model, and perform target door detection through the trained target detection model.

[0109] If a target door is detected from the third video frames of the third consecutive quantity by the trained object detection model, it is determined that there is an arriving train at the rail transit station;

[0110] Or,

[0111] Before closing the intrusion detection of the first supervision area, the detection method for intrusion personnel in the rail transit supervision area further includes:

[0112] Input the video frames intercepted from the camera video data into the trained object detection model, and perform carriage door detection through the trained object detection model.

[0113] If a carriage door is detected from the fourth video frames of the fourth consecutive quantity by the trained object detection model, it is determined that there is an arriving train at the rail transit station.

[0114] Specifically, the camera can record the rail transit station in real time to obtain a video stream. Therefore, the first camera video data can include the video before and after the train enters the station. Based on this, the third video frame is intercepted from the first camera video data, where the third video frame is earlier than the first video frame. Input multiple third video frames into the object detection model for object detection. If it is determined that a target door (specifically, the target door in the closed state) is detected according to the third video frames of the third consecutive quantity, it can be accurately determined that there is an arriving train at the rail transit station.

[0115] In this embodiment, the object detection of the target door and the detection of the opening and closing state of the target door can be performed by the same object detection model, which reduces the model training complexity and also improves the object detection efficiency. In addition, after it is determined that a target door is detected according to the third video frames of the third consecutive quantity in this embodiment, it is determined that the train has arrived, which reduces the false positive rate and improves the accuracy of arrival detection.

[0116] In another embodiment, the carriage door includes a target door and other non-target doors.

[0117] If the train includes various carriage doors such as a driver's door and passenger doors, other passenger doors can also be monitored for whether the train has arrived, not limited to the detection of the target door.

[0118] In this embodiment, the object detection model can detect not only the target door in the open state and the target door in the closed state, but also other carriage doors, such as passenger doors. Based on this, the fourth video frame is intercepted from the camera video data, where the fourth video frame is earlier than the first video frame. Input multiple fourth video frames into the object detection model for object detection. If it is determined that a carriage door is detected according to the fourth video frames of the fourth consecutive quantity, it can be accurately determined that there is an arriving train at the rail transit station.

[0119] Through the detection of the car door in this embodiment, it is possible to more accurately determine whether the train has arrived at the station, further improving the accuracy of train arrival detection.

[0120] Taking the subway as an example, during this process, misdetection of the door opening and closing status may occur. To solve this problem, this application introduces a continuous detection mechanism: only when ten consecutive frames of the car door are detected, it is determined that the subway has arrived at the station; similarly, only when ten consecutive frames of the car door are detected as open, it is determined that the car door is open, thus effectively reducing the false alarm rate. In addition, to avoid misjudging the structure similar to the car door of the passenger door between carriages as the car door during the train's operation, the discriminant logic of the carriage category is added to the target detection algorithm to improve the detection accuracy and robustness.

[0121] Through the above measures, this application can ensure the accurate monitoring of the status of the subway car door, improving the efficiency and reliability of the overall safety management.

[0122] In one embodiment, in response to the target door being in the door open state and there being a personnel intrusion in the second supervision area, a first warning is issued, including:

[0123] Input the video frame intercepted from the camera video data into the trained target detection model, and use the trained target detection model to detect the target of the intruding personnel in the second supervision area;

[0124] If it is determined that an intruding person is detected in the second supervision area based on the obtained target detection result, a first warning is issued.

[0125] Specifically, both the first supervision area and the second supervision area are beside the train track of the rail transit station. Therefore, cameras can be deployed near the track, and through this camera, the captured pictures or camera videos including the first supervision area and the second supervision area can be taken.

[0126] Of course, in a specific embodiment, the shooting angle of the camera can also be set so that the camera can capture the train track (which can capture the target door of the arriving train), the first supervision area, and the second supervision area in the same video. That is, the camera video can also include the arriving train, etc., and this application does not limit this. This can reduce the use and waste of cameras while ensuring the time consistency of the detection of the arriving train, the target door, and the supervision area.

[0127] To prevent missed shooting, this embodiment preferably shoots the camera video and transmits the camera video as camera video data to the computer device of the monitoring system.

[0128] The computer device can intercept video frames from the captured video data frame by frame, or can intercept video frames at a predetermined time interval from the captured video data. This embodiment does not limit this.

[0129] The video frames intercepted from the captured video data are sequentially input into the trained target detection model in the order of shooting. The trained target detection model can identify and detect the second supervision area without personnel intrusion and the second supervision area with personnel intrusion. That is, the second supervision areas in the above two states are both targets to be detected.

[0130] The trained target detection model performs target detection or target capture on the second supervision area without personnel intrusion and the second supervision area with personnel intrusion in the video frame, frames the target area or the candidate area containing the target, and classifies the target area or the candidate area to obtain the target detection result. Among them, the target detection result includes the target type and the calibration coordinates of the detected target. The target detection result indicates the target type of the detected target in the video frame, and the target type is the second supervision area without personnel intrusion or the second supervision area with personnel intrusion. Therefore, it can be determined whether there is personnel intrusion in the second supervision area according to the target detection result.

[0131] In a specific embodiment, the target detection model can be constructed based on the detection algorithm of the YOLO (You Only Look Once) series for example. The detection algorithm is, for example, one of YOLO V1, YOLO V5, YOLO V7, YOLO V8, etc.

[0132] In a specific embodiment, the target detection model has the function of detecting the target door in different door opening and closing states, and also has the function of detecting whether there is personnel intrusion in the second supervision area.

[0133] In a specific embodiment, the trained target detection model can perform target detection on the first supervision area without personnel intrusion and the first supervision area with personnel intrusion.

[0134] Closing the intrusion detection function for the first supervision area can be to close the output function of the second target detection result of the trained target detection model for the first supervision area. Or, closing the intrusion detection function for the first supervision area can be to shield or discard the target detection result of the target detection model output for target detection of the first supervision area, or to suspend the execution of the code for intrusion detection or alarm for the first supervision area. In this way, the monitoring system will not obtain the target detection result corresponding to the first supervision area. Therefore, during the process when the target door is in the door open state, even if the conductor or the train captain stays in the first supervision area, the alarm function will not be triggered.

[0135] In this embodiment, the target detection model can detect intrusion personnel only in the second supervision area during the opening process of the target door, and give an early warning in time to prevent misjudging the train conductor or the train attendant as an intrusion personnel, which improves the accuracy and reliability of the detection of illegal personnel intrusion in rail transit and enhances the efficiency and reliability of the overall safety management.

[0136] In one embodiment, in response to the presence of intrusion personnel in at least one of the first supervision area and the second supervision area, a second early warning is given, including:

[0137] Input the video frames intercepted from the camera video data into the trained target detection model, and use the trained target detection model to detect the intrusion personnel targets in the first supervision area and the second supervision area;

[0138] If it is determined according to the obtained target detection result that there are intrusion personnel entering at least one of the first supervision area and the second supervision area from a non-supervision area, a second early warning is given.

[0139] Specifically, both the first supervision area and the second supervision area are beside the train tracks of the rail transit station. Therefore, cameras can be arranged near the tracks, and the cameras can capture pictures or camera videos including the first supervision area and the second supervision area.

[0140] Of course, in a specific embodiment, the shooting angle of the camera can also be set so that the camera can capture the train tracks (which can capture the target door of the arriving train), the first supervision area and the second supervision area in the same video. That is, the camera video can also include the arriving train, etc., and this application does not limit this. This can reduce the use and waste of cameras, and at the same time ensure the time consistency of the detection of the arriving train, the target door, and the supervision area.

[0141] To prevent missed shooting, this embodiment preferably captures a camera video and transmits the camera video as the third camera video data to the computer device of the monitoring system.

[0142] The computer device can intercept video frames from the camera video data frame by frame, or can intercept video frames at a predetermined time interval from the camera video data. This embodiment does not limit this.

[0143] The video frames intercepted from the camera video data are input into the trained target detection model in the order of shooting. The trained target detection model can identify and detect the first supervision area without personnel intrusion, the first supervision area with personnel intrusion, the second supervision area without personnel intrusion, and the second supervision area with personnel intrusion. That is, the supervision areas in the foregoing 4 states are all targets to be detected.

[0144] The trained object detection model performs object detection or object capture on the first supervision area without personnel intrusion, the first supervision area with personnel intrusion, the second supervision area without personnel intrusion, and the second supervision area with personnel intrusion in the video frame, frames the target area or the candidate area containing the target, and classifies the target area or the candidate area to obtain the object detection result. Among them, the object detection result includes the object type and the calibrated coordinates of the detected object. The object detection result indicates the object type of the detected object in the video frame. The object type is the second supervision area without personnel intrusion or the second supervision area with personnel intrusion or the first supervision area without personnel intrusion or the first supervision area with personnel intrusion. Therefore, according to the object detection result, it can be determined whether there is personnel intrusion in the first supervision area and the second supervision area.

[0145] In a specific embodiment, within the supervision area, the system will perform intelligent personnel detection on each frame of the picture to ensure timely detection of the intrusion behavior of unauthorized personnel. When the system detects someone entering the supervision area, it will immediately issue an alarm (for example, the area in the video will flash alternately red and orange). To avoid the situation of misjudging the railway track as a human body, which may lead to false intrusion warnings. This embodiment adds a judgment mechanism on the basis of detection: the system will only trigger an intrusion warning when it detects that a human body enters the supervision area from outside the non-supervision area (such as climbing over the curtain door to enter the supervision area or coming out of the train to enter the supervision area, etc.). This measure effectively reduces the false alarm probability and ensures the safety and efficiency of the system.

[0146] In a specific embodiment, the object detection model can be constructed, for example, based on the detection algorithm of the YOLO (You Only Look Once) series. The detection algorithm is, for example, one of YOLO V1, YOLO V5, YOLO V7, YOLO V8, etc.

[0147] In a specific embodiment, the object detection model has the function of simultaneously detecting whether there is personnel intrusion in the first supervision area and the second supervision area.

[0148] In a specific embodiment, the object detection model has the function of detecting the target door in different door opening and closing states, and also has the function of simultaneously detecting whether there is personnel intrusion in the first supervision area and the second supervision area.

[0149] Turning off the intrusion detection function for the first supervision area may be turning off the output function of the trained object detection model for the object detection results of the first supervision area. Or, turning off the intrusion detection function for the first supervision area may be masking or discarding the object detection results of the object detection model output for the object detection of the first supervision area. Or, pausing the execution of the code for intrusion detection or alarm for the first supervision area, etc. This application does not limit this. In this way, the monitoring system will not obtain the object detection results corresponding to the first supervision area. Therefore, during the process when the target door is in the door open state, even if the conductor or the train captain stays in the first supervision area, the alarm function will not be triggered.

[0150] Similarly, turning on the intrusion detection function for the first supervision area may be turning on the output function of the trained object detection model for the object detection results of the first supervision area. Or, turning on the intrusion detection function for the first supervision area may be transmitting the object detection results of the object detection model output for the object detection of the first supervision area to the relevant modules of the computer device or the monitoring system. In this way, the monitoring system will obtain the object detection results corresponding to the first supervision area. Therefore, after the target door is in the door closed state, if there are people staying in the first supervision area, the alarm function will be triggered.

[0151] In a specific embodiment, the same camera can record videos for different time periods. This camera can simultaneously capture images of the track or the train on the track, the target door of the train, the first supervision area, and the second supervision area, so that the intercepted video frames can simultaneously include the target door, the first supervision area, and the second supervision area.

[0152] In this embodiment, the object detection model can detect intrusion personnel in the first supervision area and the second supervision area simultaneously after the target door is closed, and give an early warning in time.

[0153] In an embodiment, the detection of intrusion personnel in the rail transit supervision area further includes:

[0154] Obtain the first sample data, where the first sample data includes the arriving train frame images intercepted from the sample video, and the target doors in the door open state and the target doors in the door closed state are framed in the arriving train frame images;

[0155] Use the first sample data to train the to-be-trained first object detection model to obtain the trained first object detection model.

[0156] Specifically, the first sample data includes multiple frame images of arriving trains. In each frame image of an arriving train, the target doors in the door opening state and the target doors in the door closing state are framed and classified. Therefore, the first sample data can be used to train the first object detection model for detecting target doors in different states.

[0157] In addition, the performance of the trained first object detection model can also be detected through a test set.

[0158] In one embodiment, second sample data is obtained. The second sample data includes frame images of the second supervision area intercepted from a sample video. In the frame images of the second supervision area, the second supervision area with personnel intrusion and the second supervision area without personnel intrusion are framed.

[0159] The second sample data is used to train the second object detection model to be trained, and a trained second object detection model is obtained.

[0160] In one embodiment, third sample data is obtained. The third sample data includes frame images of the first supervision area and the second supervision area intercepted from a sample video. In the frame images of the first supervision area, the first supervision area with personnel intrusion and the first supervision area without personnel intrusion are framed and classified. In the frame images of the second supervision area, the second supervision area with personnel intrusion and the second supervision area without personnel intrusion are framed and classified.

[0161] The third sample data is used to train the third object detection model to be trained, and a trained third object detection model is obtained.

[0162] It should be noted that the above first object detection model, second object detection model, and third object detection model can be the same object detection model, which simultaneously has functions such as arrival detection, target door detection, target door opening and closing state detection, detection of whether there are personnel intrusions in the first supervision area, and detection of whether there are personnel intrusions in the second supervision area.

[0163] Figure 6 This is the schematic flowchart of the second embodiment of the method for detecting intruders in the rail transit supervision area in the embodiments of the present application; refer to Figure 6, in this embodiment, taking the subway as an example, camera video data is obtained, and the camera video data is intercepted frame by frame to detect whether there is a car door (target door). If a car door is detected, it indicates that there is a stopped arriving train on the track, that is, the subway has arrived at the station. In the case of the subway arriving at the station, the detection of area 1 (the first supervision area) and the detection of area 2 (the second supervision area) are started. Continuing to detect whether the car door of the car door is opened according to the intercepted frame image. If the car door is opened, it is determined that the subway door is opened, the detection of area 1 is closed, and the detection of area 2 is maintained. If it is detected that the car door is not opened or closed, it is determined that the subway door is closed, and the detection of area 1 is restarted and the detection of area 2 is maintained.

[0164] If no car door is detected, it indicates that there is no stopped arriving train on the track, that is, the subway has not arrived at the station. In the case of no subway arriving at the station, this embodiment also starts the detection of area 1 (the first supervision area) and the detection of area 2 (the second supervision area).

[0165] In this embodiment, regardless of whether the subway has arrived at the station, the intrusion detection of area 1 and area 2 is started. When the subway arrives at the station and the car door is opened, the intrusion detection of area 1 is closed, and the intrusion detection of area 2 is retained. In the case of the car door being closed, the detection of (the first supervision area) and the detection of area 2 (the second supervision area) are started.

[0166] This application significantly improves the safety monitoring effect of rail transit including the subway head area by introducing advanced computer vision technology and intelligent detection mechanisms. Under different lighting conditions, the system can accurately identify the intrusion behavior of unauthorized personnel, effectively reduce the false alarm rate, and ensure accurate alarm triggering. The flexible area division design enables the monitoring to dynamically adapt to the conductor's patrol and environmental changes, further enhancing the robustness of the system. In addition, the discrimination logic of the car body category in the target detection algorithm ensures the accurate identification of similar structures and improves the stability of the monitoring. Overall, this application effectively improves the safety and management efficiency of rail transit operation, providing a solid guarantee for the safety of passengers and vehicles.

[0167] In one embodiment, this application also provides a detection system for intrusion of personnel in the rail transit supervision area. The detection system for intrusion of personnel in the rail transit supervision area includes:

[0168] A detection control module for performing intrusion detection on the first supervision area and the second supervision area of the rail transit station. Among them, the first supervision area includes the area around the train track where the conductor is allowed to conduct on-site inspections, and the second supervision area includes the prohibited stay area around the train track except the first supervision area;

[0169] The detection control module is further configured to, in response to the presence of an arriving train at a rail transit station and the target door of the arriving train being in the door open state, turn off the intrusion detection for the first supervision area.

[0170] In one embodiment, the detection control module is further configured to, in response to the target door changing from the door open state to the door closed state, restart the intrusion detection for the first supervision area.

[0171] In one embodiment, the detection system for intruders in the rail transit supervision area further includes:

[0172] A first intrusion detection module, configured to perform a first warning in response to the target door being in the door open state and the presence of an intrusion by a person in the second supervision area;

[0173] And / or

[0174] A second intrusion detection module, configured to perform a second warning in response to the presence of an intruder in at least one of the first supervision area and the second supervision area.

[0175] In one embodiment, the detection system for intruders in the rail transit supervision area further includes: a door state detection module, configured to:

[0176] Input a video frame intercepted from the captured video data into a trained target detection model, and perform target door detection through the trained target detection model;

[0177] Determine the door open / closed state of the target door according to the obtained target detection result, where the door open / closed state includes the door open state and the door closed state.

[0178] In one embodiment, the detection control module is specifically configured to: in response to the presence of an arriving train at a rail transit station and the target door being detected as being in the door open state in a continuous first number of first video frames, turn off the intrusion detection for the first supervision area.

[0179] In one embodiment, the detection system for intruders in the rail transit supervision area further includes: an arrival detection module;

[0180] Wherein, the arrival detection module is configured to:

[0181] Input a video frame intercepted from the captured video data into a trained target detection model, and perform target door detection through the trained target detection model,

[0182] If the target door is detected by the trained target detection model in a continuous third number of third video frames, it is determined that there is an arriving train at the rail transit station;

[0183] Or,

[0184] The arrival detection module is used for:

[0185] Input the video frames intercepted from the camera video data into the trained object detection model, and perform carriage door detection through the trained object detection model.

[0186] If the carriage door is detected from the fourth consecutive number of fourth video frames through the trained object detection model, it is determined that there is an arriving train at the rail transit station.

[0187] In one embodiment, the second intrusion detection module is specifically used for:

[0188] Input the video frames intercepted from the camera video data into the trained object detection model, and perform intrusion personnel target detection on the first supervision area and the second supervision area through the trained object detection model;

[0189] If it is determined according to the obtained object detection result that there is an intrusion personnel entering at least one of the first supervision area and the second supervision area from the non-supervision area, a second warning is given.

[0190] Reference Figure 7 In a specific embodiment, the present application further provides a detection system for intrusion personnel in the rail transit supervision area. The detection system for intrusion personnel in the rail transit supervision area includes: a detection control module 100, an arrival detection module 200, a door state detection module 300, a first intrusion detection module 400, and a second intrusion detection module 500.

[0191] For the functions of the detection control module 100, the arrival detection module 200, the door state detection module 300, the first intrusion detection module 400, and the second intrusion detection module 500, refer to the above description, and details are not described here again.

[0192] For the working principle of the detection system for intrusion personnel in the rail transit supervision area of the present application, refer to the description of the detection method for intrusion personnel in the rail transit supervision area above, and details are not described here again.

[0193] The present application can effectively improve the intelligence and accuracy of safety monitoring in rail transit, such as the front area of a subway train, and particularly focuses on the detection of the behavior of non-train crew members climbing over the platform screen door. By combining advanced computer vision technology, the system can monitor the position of the human detection center point under different lighting conditions and changes in the position of the train head, so as to accurately judge the intrusion behavior of unauthorized personnel. This technology significantly reduces false alarms and interference by reasonably dividing the intrusion detection area into Area 1 and Area 2 and setting an effective train crew patrol detection logic, and significantly enhances the overall monitoring effect. Compared with the existing methods, the present application has the following advantages:

[0194] Precise Intrusion Detection and Intelligent Monitoring:

[0195] By introducing a continuous detection mechanism, this application significantly reduces the false alarm rate, ensuring that intrusion alarms are only triggered under confirmed conditions, thereby improving the accuracy of detection. In addition, the system automatically identifies the arrival of the train and the door status, achieving automated management. This intelligent safety monitoring level reduces the need for manual intervention, thereby improving operational efficiency and safety.

[0196] Flexible Area Division and Strong Robustness:

[0197] The system divides the intrusion detection area into Area 1 and Area 2 to flexibly adapt to different monitoring conditions and optimize the relationship between the conductor's patrol and intrusion detection. At the same time, the discriminant logic of the carriage category is introduced into the target detection algorithm to effectively avoid misjudging other structures with similar shapes as the train door. This design significantly enhances the stability and reliability of monitoring, ensuring effective operation in complex environments.

[0198] Improve Safety Management Efficiency:

[0199] This application not only enhances the detection ability of unauthorized personnel intrusion but also can effectively reduce false alarms during the conductor's normal patrol, ensuring the safety and smoothness of subway operation. Through these innovations, the system can significantly improve safety management efficiency in actual operation and effectively guarantee the safety of passengers and vehicles.

[0200] Figure 8 It is a schematic structural diagram of a computer device provided by an embodiment of this application. The computer device 700 may vary greatly due to configuration or performance differences and may include one or more processors (central processing units, CPUs) 710 (for example, one or more processors) and a memory 720, and one or more storage media 730 (for example, one or more mass storage devices) for storing application programs 733 or data 732. Among them, the memory 720 and the storage media 730 can be transient storage or persistent storage. The program stored in the storage media 730 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the computer device 700. Further, the processor 710 can be set to communicate with the storage media 730 and execute a series of instruction operations in the storage media 730 on the computer device 700.

[0201] The computer device 700 may also include one or more power supplies 740, one or more wired or wireless network interfaces 750, one or more input / output interfaces 760, and / or one or more operating systems 731, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that Figure 8 The computer device structure shown does not constitute a limitation on the computer device and may include more or fewer components than shown, or combine certain components, or have a different component arrangement.

[0202] This application also provides a computer device. The computer device includes a memory and a processor. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the processor, the processor is caused to execute the steps of the method for detecting intruders in the rail transit supervision area in the above-mentioned various embodiments. This application also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions run on a computer, the computer is caused to execute the steps of the method for detecting intruders in the rail transit supervision area.

[0203] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0204] If the integrated unit is implemented in the form of 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 this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0205] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting intruders in a rail transit supervision area, characterized in that: The method for detecting a person intruding into a rail transit supervision area includes: Performing intrusion detection on a first supervision area and a second supervision area of a rail transit station, wherein the first supervision area includes an area around the train tracks where train crew members are allowed to patrol the station, and the second supervision area includes a prohibited area around the train tracks other than the first supervision area; In response to a train arriving at the rail transit station and a target door of the arriving train being in an open state, disabling intrusion detection on the first supervision area; In response to the target vehicle door changing from an open state to a closed state, re-enabling intrusion detection on the first supervision area; In response to the target vehicle door being in an open state and a person intruding in the second supervision area, a first warning is issued; and / or, in response to the target vehicle door being in a closed state and a person intruding in at least one of the first supervision area and the second supervision area, a second warning is issued.

2. The method for detecting intruders in a rail transit supervision area according to claim 1, characterized in that: The method for detecting a person intruding into a rail transit supervision area further includes: Get the door switch status sent by the arrival station list; or, Computer vision technology is used to monitor the door switch status of arriving trains.

3. The method for detecting intruders in a rail transit supervision area according to claim 1, characterized in that: In response to the target vehicle door being in an open state and a person intruding into the second supervision area, issuing a first warning includes: In response to the target vehicle door being in an open state, a person intruding into the second supervision area, and detecting that the intruder enters the second supervision area from outside the non-supervision area, a first warning is issued.

4. The method for detecting intruders in a rail transit supervision area according to claim 1, characterized in that: Before shutting down the intrusion detection on the first supervision area, the method for detecting an intruder in the rail transit supervision area further includes: Inputting the video frames captured from the camera video data into the trained target detection model, and performing target door detection using the trained target detection model; The door switch state of the target door is determined according to the obtained target detection result, wherein the door switch state includes a door open state and a door closed state.

5. The method for detecting intruders in rail transit supervision areas according to claim 4, characterized in that: In response to a train arriving at the rail transit station and a target door of the arriving train being in an open state, disabling intrusion detection on the first supervision area includes: In response to the presence of an arriving train at the rail transit station and the detection of the target doors being in an open state from a first number of consecutive first video frames, intrusion detection on the first supervision area is closed.

6. The method for detecting intruders in rail transit supervision areas according to claim 1, characterized in that: Before shutting down the intrusion detection on the first supervision area, the method for detecting an intruder in the rail transit supervision area further includes: Input the video frames captured from the camera video data into the trained target detection model, and perform target door detection through the trained target detection model. If the target door is detected from a third number of consecutive third video frames by the trained target detection model, it is determined that there is an arriving train at the rail transit station; or, Before shutting down the intrusion detection on the first supervision area, the method for detecting an intruder in the rail transit supervision area further includes: The video frames captured from the camera video data are input into the trained object detection model, and the door detection is performed by the trained object detection model. If a carriage door is detected from a fourth number of consecutive fourth video frames by the trained target detection model, it is determined that there is an arriving train at the rail transit station.

7. The method for detecting intruders in rail transit supervision areas according to claim 1, characterized in that: The step of issuing a second early warning in response to the presence of an intruder in at least one of the first supervision area and the second supervision area includes: Inputting video frames captured from the camera video data into a trained target detection model, and performing intruder target detection on the first and second supervision areas using the trained target detection model; If it is determined according to the obtained target detection result that an intruder has entered from the non-supervised area into at least one of the first and second supervised areas, a second early warning is issued.

8. A system for detecting intruders in rail transit supervision areas, characterized in that: The rail transit supervision area intruder detection system includes: a detection control module, configured to perform intrusion detection on a first supervision area and a second supervision area of a rail transit station, wherein the first supervision area includes an area around the train tracks where train crew members are allowed to patrol the station, and the second supervision area includes a prohibited area around the train tracks other than the first supervision area; The detection control module is further configured to disable intrusion detection on the first supervision area in response to the presence of an arriving train at the rail transit station and the target door of the arriving train being in an open state; The detection control module is further configured to re-enable intrusion detection of the first supervision area in response to the target vehicle door changing from an open state to a closed state; The first intrusion detection module is used to issue a first warning in response to the target vehicle door being in an open state and the presence of a person intruding in the second supervision area; and / or the second intrusion detection module is used to issue a second warning in response to the target vehicle door being in a closed state and the presence of an intruder in at least one of the first supervision area and the second supervision area.

9. A computer device, characterized in that: The computer device includes: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the computer device to execute the method for detecting a person intruding into a rail transit supervision area according to any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the method for detecting a person intruding into a rail transit supervision area according to any one of claims 1 to 7 is implemented.

Images