Subway field section multi-scene linkage safety management and control system and method

Abstract

The invention relates to the technical field of metro field data processing, and discloses a metro field multi-scene linkage safety management and control system and method, and the system comprises an access control module, a monitoring module, an illumination module, a passenger flow prediction module, an abnormality detection module, and a security strategy module. Wherein the access control module records real-time passenger flow volume; the monitoring module is used for acquiring a monitoring video of a subway field section; the lighting module is used for controlling the working state of lighting equipment of the subway field section; the passenger flow prediction module is used for calculating a passenger flow prediction value; the anomaly detection module is used for carrying out anomaly detection; the security strategy module selects a security strategy based on the passenger flow volume predicted value and the abnormal alarm signal; according to the invention, through multi-system linkage, intelligent passenger flow management, comprehensive anomaly detection and dynamic illumination control, the safety management level of the metro field is comprehensively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of subway yard data processing, and in particular to a subway yard multi-scenario linkage safety management and control system and method. Background Art

[0002] At present, the subway systems in many cities at home and abroad have realized multiple functions such as automated operation, intelligent dispatching, and passenger information services. For example, by introducing advanced computer technology and communication technology, functions such as automatic train control and automatic station services are realized, reducing the error rate of manual operation; dynamically adjusting train operation plans based on real-time passenger flow data, weather changes and other factors, optimizing train departure intervals, and reducing passenger waiting time; equipped with video surveillance, fire alarms, emergency communications and other safety facilities, improving the level of safety management, etc.

[0003] Although the subway system has made significant progress in intelligence and informatization, there are still some shortcomings in safety management and control, which are mainly reflected in the following aspects: At present, the safety management and control system of subway stations is often composed of multiple independent subsystems, such as video surveillance system, access control system, fire alarm system, etc., and there is a lack of effective linkage mechanism between the subsystems. This independent operation mode not only increases the complexity and operation and maintenance cost of the system, but also makes it difficult to achieve rapid response and coordinated disposal when emergencies occur, affecting the overall safety management level. Although some subway systems have introduced passenger flow prediction models, the accuracy of existing passenger flow predictions still needs to be improved due to reasons such as a single data source and immature algorithms. In addition, the existing passenger flow management measures are mostly post-event responses, lack of foresight, and cannot effectively prevent safety accidents such as crowding and trampling caused by a surge in passenger flow. At present, the lighting systems of most subway stations still operate in a fixed mode and cannot dynamically adjust the lighting brightness according to actual needs. This not only wastes energy, but also at night or under low light conditions, the lower lighting brightness may affect passengers' sight and increase safety hazards.

[0004] For example, a Chinese patent application with publication number CN115600996A discloses a subway security integration platform and a subway security resource point calibration method, wherein the subway security integration platform includes: a security front-end subsystem, including a display module and an input module; a security back-end subsystem, including a technical defense subsystem module and a configuration management module, wherein the technical defense subsystem module includes a security resource unit, a communication unit, a processing unit and a storage unit; the security resource unit is connected to the processing unit through the communication unit, and the collected security data is transmitted to the processing unit, and the storage unit is used to store the security data before and after processing; the configuration management module is used to mark and manage the subway lines, processing units, trains and security resource units according to preset rules, establish a linkage relationship and feedback to the processing unit.

[0005] The above patent has the problems raised by this background technology: the passenger flow management measures are post-event responses, lack foresight, and cannot effectively prevent safety accidents such as crowding and trampling caused by the surge in passenger flow; the lighting system still operates in a fixed mode and cannot dynamically adjust the lighting brightness according to actual needs.

[0006] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention

[0007] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a multi-scenario linkage safety management and control system and method for a subway station. The present invention comprehensively improves the safety management level of subway stations through multi-system linkage, intelligent passenger flow management, comprehensive anomaly detection, and dynamic lighting control.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] On the one hand, the present invention provides a multi-scenario linkage security management and control system for a subway station, including an access control module, a monitoring module, a lighting module, a passenger flow prediction module, an anomaly detection module, and a security strategy module; wherein:

[0010] The access control module is used to control the entry and exit of passengers and record the real-time passenger flow;

[0011] The monitoring module is used to obtain the monitoring video of the subway station and provide the location of each monitoring video;

[0012] The lighting module is used to control the working status of the lighting equipment in the subway station and monitor the status of each lighting equipment;

[0013] The passenger flow prediction module calculates the passenger flow prediction value based on environmental information and historical passenger flow data;

[0014] The anomaly detection module performs anomaly detection based on the surveillance video of the subway station and generates an abnormal alarm signal;

[0015] The security strategy module selects a security strategy based on the passenger flow prediction value and the abnormal alarm signal; the access control module, the lighting module, and the abnormal detection module respectively adjust the access control strategy, the lighting control strategy, and the abnormal detection strategy based on the security strategy.

[0016] As a preferred solution of the multi-scenario linkage safety control system of the subway station described in the present invention, wherein: the access control module includes an access strategy unit and a counting unit; wherein the access strategy unit is configured with an access control strategy; the counting unit is used to record the passenger flow in the station in real time;

[0017] The access control strategy includes a first access strategy and a second access strategy; wherein the first access strategy includes: the number of entrances open is , the open quantity for export is , the number of open ticket inspection channels is ; The first entry and exit strategy also includes The number of open entrances, The number of open exits, The number of the open ticket inspection channel;

[0018] The second entry and exit strategy includes: the number of entrances opened is , the open quantity for export is , the number of open ticket inspection channels is ; The second entry and exit strategy also includes The number of open entrances, The number of open exits, The number of the open ticket inspection channel.

[0019] As a preferred solution of the multi-scenario linkage security management and control system of the subway section described in the present invention, the monitoring module includes a camera unit and a map unit; wherein the camera unit is used to obtain monitoring videos of the subway section from monitoring cameras distributed in the subway section; the map unit is used to provide an electronic map of the subway section; the electronic map provides the positioning of the monitoring camera corresponding to each monitoring video and the positioning of each sensor in the subway section.

[0020] As a preferred solution of the multi-scenario linkage safety management and control system of the subway section of the present invention, wherein: the lighting module includes a monitoring unit and a control unit; wherein the monitoring unit is used to monitor the working status of each lighting device in real time; the control unit is configured with a lighting control strategy, and the lighting control strategy includes the lighting brightness of each lighting device in the subway section; the monitoring unit is configured with standard working parameters of each lighting device corresponding to the lighting control strategy, and if the working parameters of any lighting device are inconsistent with the corresponding standard working parameters, lighting abnormality information is sent to the map unit; the map unit marks the corresponding position of the lighting abnormality information in the electronic map;

[0021] The lighting control strategy includes a first lighting strategy, a second lighting strategy, and a third lighting strategy; wherein, in the first lighting strategy, the lighting brightness of each lighting device in the subway section is the first brightness of the corresponding device; and the second lighting strategy includes the lighting brightness of each lighting device in the subway section being the second brightness of the corresponding device.

[0022] As a preferred solution of the multi-scenario linkage safety management and control system of the subway section of the present invention, the passenger flow prediction module includes a data acquisition unit and a prediction model unit; wherein the data acquisition unit is used to collect environmental information, obtain historical passenger flow data from the counting unit, and organize the historical passenger flow data into a time series; wherein the environmental information includes date information, time point, temperature, precipitation, and passenger flow of adjacent stations; wherein the date information includes holiday tags and week tags; the historical passenger flow data includes real-time passenger flow at the previous m time points, where m is a positive integer;

[0023] The prediction model unit is configured with a prediction model, which is used to calculate the passenger flow prediction value; the input of the prediction model includes the feature vector of the environmental information and the time series of historical passenger flow data, and the output is the passenger flow prediction value within the next n minutes, where n is a positive integer.

[0024] As a preferred solution of the multi-scenario linkage safety management and control system for a subway section described in the present invention, wherein: the abnormality detection module includes a detection control unit; the detection control unit is configured with an abnormality detection strategy; the abnormality detection module also includes a first detection unit, which is used to perform grounding pole status detection, air-conditioning cover detection, fire smoke detection, and turnout jam detection through sensors; when any abnormal information of grounding pole abnormality, air-conditioning cover not closed correctly, fire smoke, and turnout jam is detected, the first detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the area coordinates corresponding to the abnormal alarm signal at the corresponding position in the electronic map.

[0025] As a preferred solution of the multi-scene linkage safety management and control system of the subway section of the present invention, the abnormal detection module also includes a second detection unit, which performs object left behind detection, animal intrusion detection, and train behavior detection based on the surveillance video of the subway section; wherein, the method of performing object left behind detection and animal intrusion detection is as follows: collecting scene images that do not contain left behind objects and wild animals in the surveillance video, and establishing a scene background model; at fixed detection cycles, intercepting scene images from the surveillance video, and performing foreground detection, extracting foreground images containing foreground objects; processing the foreground images through an image recognition model, and identifying and classifying foreground objects in the foreground images;

[0026] The method of performing train behavior detection is as follows: at every fixed detection period, a monitoring image containing a train is intercepted from the monitoring video; a target detection algorithm is used to detect the train in the monitoring image; a multi-target tracking algorithm is used to continuously track the monitored train to obtain the train's motion trajectory, including the position and timestamp; the speed and dwell time of the train are calculated based on the train's motion trajectory; and train behavior detection is performed based on the speed and dwell time of the train;

[0027] When any abnormal information such as abandoned items, wild animals, and abnormal train behavior is detected, the second detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the area coordinates corresponding to the abnormal alarm signal at the corresponding position in the electronic map.

[0028] As a preferred solution of the multi-scene linkage security management and control system for a subway section described in the present invention, the abnormal detection module also includes a third detection unit, which performs illegal intrusion detection based on the surveillance video of the subway section; the method is as follows: at fixed detection cycles, a surveillance image is captured from the surveillance video of each security control area, and portrait detection is performed on the surveillance image; if a portrait is detected in the surveillance image, face recognition is performed on the portrait; if the face recognition result is an authorized manager, no illegal intrusion occurs; otherwise, illegal intrusion occurs; when an illegal intrusion is detected, the third detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the corresponding position of the area coordinates corresponding to the abnormal alarm signal in the electronic map.

[0029] As a preferred solution of the multi-scenario linkage safety management and control system of the subway section of the present invention, the abnormal detection strategy includes a first detection strategy, a second detection strategy, and a third detection strategy; wherein the first detection strategy includes: the detection cycle of the second detection unit is , the detection cycle of the third detection unit is The second detection strategy includes: the detection cycle of the second detection unit is , the detection cycle of the third detection unit is .

[0030] As a preferred solution of the multi-scenario linkage security management and control system for a subway section described in the present invention, wherein: the security strategy includes a first security strategy, a second security strategy, and a third security strategy; the security strategy module includes a strategy unit and an instruction unit; wherein the strategy unit is configured with a passenger flow threshold for selecting a security strategy; the strategy unit obtains a passenger flow prediction value from the prediction model unit, and if the passenger flow prediction value is not higher than the passenger flow threshold, the strategy unit selects the first security strategy; the instruction unit generates a first security instruction based on the first security strategy, and sends it to the entry and exit strategy unit, the control unit, and the detection control unit respectively; the entry and exit strategy unit responds to the first security instruction to trigger the first entry and exit strategy; the control unit responds to the first security instruction to trigger the first lighting strategy; the detection control unit responds to the first security instruction to trigger the first detection strategy;

[0031] If the predicted passenger flow value is higher than the passenger flow threshold, the strategy unit selects the second security strategy; the instruction unit generates a second security instruction based on the second security strategy, and sends it to the entry and exit strategy unit, the control unit, and the detection control unit respectively; the entry and exit strategy unit responds to the second security instruction and triggers the second entry and exit strategy; the control unit responds to the second security instruction and triggers the second lighting strategy; the detection control unit responds to the second security instruction and triggers the second detection strategy.

[0032] As a preferred solution of the multi-scenario linkage security management and control system for a subway section described in the present invention, if the strategy unit receives an abnormal alarm signal, a third security strategy is triggered; the instruction unit generates a third security instruction based on the third security strategy, and sends it to the control unit and the detection control unit respectively; the control unit responds to the third security instruction to trigger a third lighting strategy; the third lighting strategy is specifically as follows: the coordinates of the area corresponding to the abnormal alarm signal are obtained through an electronic map, and the lighting brightness of each lighting device configured in the area corresponding to the abnormal alarm signal is set to a preset rated brightness; the detection control unit responds to the third security instruction to trigger a third detection strategy; the third detection strategy is specifically as follows: the monitoring image of the corresponding area is collected at the rated brightness of the lighting device, and the abnormal detection is re-performed through the corresponding detection unit; if the corresponding detection unit sends an abnormal alarm signal, the abnormal alarm signal is pushed to the management personnel.

[0033] In a second aspect, the present invention provides a multi-scenario linkage safety management and control method for a subway station, comprising the following steps: S1: collecting environmental information and continuously recording real-time passenger flow;

[0034] S2: Encode environmental information into feature vectors and organize the continuously recorded real-time passenger flow into a time series of historical passenger flow data;

[0035] S3: Calculate the passenger flow forecast value based on the feature vector of environmental information and the time series of historical passenger flow data;

[0036] S4: Selecting a security strategy based on the passenger flow prediction value;

[0037] S5: Set access control strategy, lighting control strategy and anomaly detection strategy based on security strategy;

[0038] S6: Perform multi-scenario linkage security management and control in subway stations based on the set access control strategy, lighting control strategy and anomaly detection strategy.

[0039] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0040] By introducing an advanced passenger flow prediction module, the passenger flow in the short term in the future can be accurately predicted. This helps to take diversion measures in advance, reasonably arrange entrances and exits and temporary security inspection channels, and effectively prevent safety accidents such as congestion and trampling caused by a surge in passenger flow. According to the passenger flow prediction results, the appropriate security strategy is automatically selected, and the access control strategy, lighting control strategy and anomaly detection strategy are dynamically adjusted to ensure the best security protection under different passenger flow conditions.

[0041] It integrates multiple anomaly detection methods, covers all key parts of the subway station, and can comprehensively and accurately identify various potential safety hazards. By using advanced image recognition technology and target detection algorithms, it automatically identifies abnormal behaviors and generates alarm signals, reducing the workload of manual monitoring and improving the accuracy and timeliness of anomaly detection.

[0042] The brightness of lighting equipment is automatically adjusted according to passenger flow forecast results and abnormal alarm signals; when the system detects an abnormal situation, it can immediately adjust the lighting equipment in the abnormal area to the preset rated brightness, providing better visual conditions for security personnel and passengers, while enabling monitoring equipment to capture clearer monitoring images, facilitating subsequent emergency handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0044] Figure 1 A schematic diagram of the structure of the multi-scenario linkage safety management and control system for a subway station provided by the present invention;

[0045] Figure 2A flow chart of the multi-scenario linkage safety control method for a subway station provided by the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0047] Example 1

[0048] This embodiment introduces a multi-scenario linkage safety control system for a subway station. Figure 1 The system includes access control module, monitoring module, lighting module, passenger flow prediction module, anomaly detection module, and security strategy module; among which:

[0049] The access control module is used to control the entry and exit of passengers and record the real-time passenger flow;

[0050] The access control module includes an access strategy unit and a counting unit; wherein the access strategy unit is configured with an access control strategy; the counting unit is used to record the passenger flow in the station in real time;

[0051] The access control strategy includes a first access strategy and a second access strategy; wherein the first access strategy includes: the number of entrances open is , the open quantity for export is , the number of open ticket inspection channels is ; The first entry and exit strategy also includes The number of open entrances, The number of open exits, The number of the open ticket inspection channel;

[0052] The second entry and exit strategy includes: the number of entrances opened is , the open quantity for export is , the number of open ticket inspection channels is ; The second entry and exit strategy also includes The number of open entrances, The number of open exits, The number of the open ticket inspection channel; among them, , are all positive integers and Less than ; , are positive integers and Less than ; , are positive integers and Less than ;

[0053] The monitoring module is used to obtain the monitoring video of the subway station and provide the location of each monitoring video;

[0054] The monitoring module includes a camera unit and a map unit; wherein: the camera unit is used to obtain monitoring videos of the subway section from monitoring cameras distributed in the subway section; the map unit is used to provide an electronic map of the subway section; the electronic map provides the location of the monitoring camera corresponding to each monitoring video and the location of each sensor in the subway section;

[0055] The lighting module is used to control the working status of the lighting equipment in the subway station and monitor the status of each lighting equipment;

[0056] The lighting module includes a monitoring unit and a control unit; wherein the monitoring unit is used to monitor the working state of each lighting device in real time; the control unit is configured with a lighting control strategy, and the lighting control strategy includes the lighting brightness of each lighting device in the subway section; the monitoring unit is configured with standard working parameters of each lighting device corresponding to the lighting control strategy, and if the working parameters of any lighting device are inconsistent with the corresponding standard working parameters, lighting abnormality information is sent to the map unit; the map unit marks the corresponding position of the lighting abnormality information in the electronic map;

[0057] The lighting control strategy includes a first lighting strategy, a second lighting strategy, and a third lighting strategy; wherein, in the first lighting strategy, the lighting brightness of each lighting device in the subway section is the first brightness of the corresponding device; the second lighting strategy includes that the lighting brightness of each lighting device in the subway section is the second brightness of the corresponding device; for any lighting device, the second brightness is greater than the first brightness;

[0058] The passenger flow prediction module calculates the passenger flow prediction value based on environmental information and historical passenger flow data;

[0059] The passenger flow prediction module includes a data collection unit and a prediction model unit; wherein the data collection unit is used to collect environmental information, obtain historical passenger flow data from the counting unit, and organize the historical passenger flow data into a time series; wherein the environmental information includes date information, time point, temperature, precipitation, and passenger flow of adjacent stations; wherein the date information includes holiday labels and week labels; for example, the holiday label includes whether it is a holiday or not, and the week label is Wednesday; the time point indicates the time point when the passenger flow prediction value is calculated; the temperature and precipitation are the temperature and precipitation of the day respectively; the passenger flow of the adjacent stations includes the passenger flow of the previous subway station and the next subway station; the historical passenger flow data includes the real-time passenger flow of the first m time points, where m is a positive integer;

[0060] The prediction model unit is configured with a prediction model, and the prediction model is used to calculate the passenger flow prediction value; the input of the prediction model includes the feature vector of the environmental information and the time series of historical passenger flow data, and the output is the passenger flow prediction value in the next n minutes, where n is a positive integer;

[0061] The prediction model includes a first input layer, a second input layer, a first processing layer, a second processing layer, a fusion layer, a fully connected layer, and an output layer; wherein: the first input layer is used to input the time series of historical passenger flow data; the second input layer is used to input the feature vector of environmental information; the first processing layer is used to extract time series features from the time series of historical passenger flow data; the first processing layer can use an LSTM layer to process time series data and capture the long-term dependencies of the time series; the second processing layer is used to extract non-time series features from the feature vector of environmental information; the second processing layer can use a fully connected layer to learn the nonlinear relationship between input features and convert the non-time series data contained in environmental factors into a form suitable for further processing; the fusion layer is used to merge time series features and non-time series features and output fusion features; the fusion layer splices features from two different sources together so that the model can simultaneously utilize information from time series data and non-time series data; the fully connected layer is used to further process the fusion features and output high-level features; the output layer is used to map the high-level features into passenger flow prediction values ​​within the next n minutes.

[0062] When training the prediction model, first collect historical passenger flow data and environmental information, then perform data preprocessing, including filling or deleting missing data, and normalize the data to scale the collected data to between 0 and 1 for easy model training. Divide the normalized data into training set, validation set, and test set. Use the training set to train the prediction model, and the validation set to adjust the hyperparameters. Use the test set to evaluate the model performance, and use indicators such as mean square error and mean absolute error to evaluate the prediction accuracy of the model. When the performance of the prediction model meets the standard and the training is completed, the trained model will be deployed and applied to predict future passenger flow in real time. Through the above steps, an efficient and accurate passenger flow prediction model can be built to provide strong support for the safety management and control of subway stations.

[0063] The anomaly detection module performs anomaly detection based on the surveillance video of the subway station and generates an abnormal alarm signal;

[0064] The abnormality detection module includes a detection control unit; the detection control unit is configured with an abnormality detection strategy; the abnormality detection module also includes a first detection unit, which is used to perform grounding rod state detection, air conditioning cover plate detection, fire smoke detection, and turnout jam detection through sensors; wherein, the grounding rod state detection is performed in the following manner: the resistance value of the corresponding position is measured by a resistance sensor installed at the connection point between the grounding clamp and the rod body, and the current value of the corresponding position is measured by a current sensor installed at the track contact point; based on the resistance value and the current value, it is determined whether the grounding rod is abnormal; the subway train and the power supply system need to be safely grounded through the grounding rod to ensure the personal safety of the staff during maintenance operations. By performing grounding rod state detection, electrical safety can be ensured and electric shock accidents can be avoided. The air conditioning cover plate detection is performed in the following manner: the locking pressure is measured by a pressure sensor installed at the contact point between the locking hook and the cover plate; based on the locking pressure, it is determined whether the air conditioning cover plate is correctly closed; a mechanical locking mechanism is used in conjunction with an electronic detection method to ensure that the air conditioning cover plate is in a safe closed state, which can ensure the normal operation of the subway train air conditioning system. If the air conditioning cover plate is not closed correctly, foreign matter may enter the air conditioning system, causing safety hazards. The method for fire smoke detection is to detect the smoke concentration in the subway station section through a smoke sensor, and judge whether fire smoke occurs based on the smoke concentration; the method for turnout jam detection is as follows: measure the displacement and driving pressure of the turnout through a displacement sensor and a pressure sensor installed on the turnout switch; judge whether the turnout jam occurs based on the displacement and driving pressure of the turnout; when any abnormal information of grounding rod abnormality, air-conditioning cover not closed correctly, fire smoke occurs, and turnout jam is detected, the first detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the corresponding position of the area coordinates corresponding to the abnormal alarm signal in the electronic map.

[0065] The anomaly detection module also includes a second detection unit, which performs object left behind detection, animal intrusion detection, and train behavior detection based on the surveillance video of the subway station; wherein, the method for performing object left behind detection and animal intrusion detection is as follows: scene images that do not contain left behind objects and wild animals in the surveillance video are collected to establish a scene background model; a Gaussian mixture model or other background modeling methods can be used to establish a background model for each scene; at fixed detection intervals, scene images are captured from the surveillance video, and foreground detection is performed to extract foreground images containing foreground objects; the foreground objects refer to objects in the scene images that do not conform to the background model, and foreground detection and foreground image extraction can be performed by background subtraction; the foreground images are processed by an image recognition model to identify and classify foreground objects in the foreground images; convolutional neural networks can be trained for common left behind objects and common wild animals as the image recognition models, respectively, to achieve classification and recognition of foreground objects.

[0066] The surveillance videos for object left behind detection include those at the track area, station entrances and exits, waiting halls, stairs, etc.; the surveillance videos for animal intrusion detection are those at the open-air track. Object left behind detection and animal intrusion detection can prevent safety accidents caused by objects left behind, and at the same time, timely detect the entry of wild animals into the track to prevent operation accidents.

[0067] The method of performing train behavior detection is as follows: at fixed detection cycles, a monitoring image containing a train is captured from the monitoring video; a target detection algorithm is used to detect the train in the monitoring image, such as a target detection algorithm such as YOLOv5 or SSD; a multi-target tracking algorithm is used to continuously track the monitored train to obtain the train's motion trajectory, including the position and timestamp. For example, tracking algorithms such as DeepSORT and FairMOT can track and record the train's motion trajectory. In each frame of the monitoring video, the center position and timestamp of the train are recorded. This information can be stored in a dictionary or list, where the timestamp can be the frame number of the video or the actual time point; the speed and dwell time of the train are calculated based on the train's motion trajectory; train behavior detection is performed based on the speed and dwell time of the train; when the train is detected to be speeding or abnormally stopped, the train behavior is abnormal; for example, a speed threshold of the train is set, and when the train speed is detected to exceed the speed threshold, it is considered to be speeding; a dwell time threshold of the train is set, and when the train dwell time is detected to exceed the dwell time threshold, it is considered to be abnormally stopped;

[0068] When any abnormal information such as abandoned items, wild animals, and abnormal train behavior is detected, the second detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the area coordinates corresponding to the abnormal alarm signal at the corresponding position in the electronic map.

[0069] The abnormal detection module also includes a third detection unit, which performs illegal intrusion detection based on the surveillance video of the subway section; the method is as follows: at fixed detection cycles, a surveillance image is captured from the surveillance video of each security control area, and a portrait detection is performed on the surveillance image; if a portrait is detected in the surveillance image, the face recognition is performed on the portrait; if the face recognition result is an authorized manager, no illegal intrusion occurs; otherwise, an illegal intrusion occurs; when an illegal intrusion is detected, the third detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the corresponding position of the area coordinates corresponding to the abnormal alarm signal in the electronic map. The third detection unit is mainly used for the anti-theft and anti-destruction monitoring management of personal and property safety and production bases in the control center, train section and parking lot to prevent unidentified persons from entering.

[0070] The anomaly detection strategy includes a first detection strategy, a second detection strategy, and a third detection strategy; wherein the first detection strategy includes: the detection cycle of the second detection unit is , the detection cycle of the third detection unit is The second detection strategy includes: the detection cycle of the second detection unit is , the detection cycle of the third detection unit is ;in, Greater than , Greater than .

[0071] The security strategy module selects a security strategy based on the passenger flow prediction value and the abnormal alarm signal; the access control module, the lighting module, and the abnormal detection module respectively adjust the access control strategy, the lighting control strategy, and the abnormal detection strategy based on the security strategy.

[0072] The security strategy includes a first security strategy, a second security strategy, and a third security strategy; the security strategy module includes a strategy unit and an instruction unit; wherein the strategy unit is configured with a passenger flow threshold for selecting a security strategy; the strategy unit obtains a passenger flow prediction value from the prediction model unit, and if the passenger flow prediction value is not higher than the passenger flow threshold, the strategy unit selects the first security strategy; the instruction unit generates a first security instruction based on the first security strategy, and sends it to the entry and exit strategy unit, the control unit, and the detection control unit respectively; the entry and exit strategy unit responds to the first security instruction and triggers the first entry and exit strategy; the control unit responds to the first security instruction and triggers the first lighting strategy; the detection control unit responds to the first security instruction and triggers the first detection strategy;

[0073] If the passenger flow prediction value is higher than the passenger flow threshold, the strategy unit selects the second security strategy; the instruction unit generates a second security instruction based on the second security strategy, and sends it to the entry and exit strategy unit, the control unit, and the detection control unit respectively; the entry and exit strategy unit responds to the second security instruction and triggers the second entry and exit strategy; the control unit responds to the second security instruction and triggers the second lighting strategy; the detection control unit responds to the second security instruction and triggers the second detection strategy;

[0074] Under the control of the first security strategy and the second security strategy selected based on passenger flow prediction, the system can automatically adjust the lighting brightness of the subway section. During low-peak passenger flow periods, the lighting brightness is appropriately reduced to save energy; during peak passenger flow periods, the lighting brightness is automatically increased, especially the lighting brightness of entrances and exits and transfer areas, which can improve passenger visibility and safety; at the same time, according to the passenger flow prediction results, diversion measures can be taken in advance, and entrances and exits and temporary security inspection channels can be added the day before the peak passenger flow period to prevent congestion. In addition, when the passenger flow prediction value exceeds the preset passenger flow threshold, the possibility of abnormal conditions in the subway section is high. The system can automatically increase the monitoring frequency and analysis intensity of the second detection unit and the third detection unit by adjusting the detection cycle to avoid missing abnormal conditions.

[0075] If the strategy unit receives an abnormal alarm signal, the third security strategy is triggered; the instruction unit generates a third security instruction based on the third security strategy and sends it to the control unit and the detection control unit respectively; the control unit responds to the third security instruction and triggers the third lighting strategy; the third lighting strategy is as follows: the coordinates of the area corresponding to the abnormal alarm signal are obtained through the electronic map, and the lighting brightness of each lighting device configured in the area corresponding to the abnormal alarm signal is set to the preset rated brightness; the preset rated brightness can provide better visual conditions for security personnel and passengers, and enable the monitoring equipment to capture clearer monitoring images. The detection control unit responds to the third security instruction and triggers the third detection strategy; the third detection strategy is as follows: the monitoring image of the corresponding area is collected at the rated brightness of the lighting device, and the abnormal detection is re-performed through the corresponding detection unit (the first detection unit or the second detection unit or the third detection unit); if the corresponding detection unit sends an abnormal alarm signal, the abnormal alarm signal is pushed to the management personnel. The area where the abnormal situation is detected is re-detected at a more suitable lighting brightness to avoid misjudgment of the abnormal situation; if the result of the secondary detection is still abnormal, it is determined that the area does have an abnormal situation, and the abnormal information needs to be pushed to the corresponding management personnel in a timely manner.

[0076] Example 2

[0077] This embodiment is the second embodiment of the present invention; it is based on the same inventive concept as embodiment 1, Figure 2 This embodiment introduces a multi-scenario linkage safety management and control method for a subway station, including the following steps:

[0078] S1: Collect environmental information and continuously record real-time passenger flow; the environmental information includes date information, time point, temperature, precipitation, and passenger flow of adjacent stations;

[0079] S2: Encode environmental information into feature vectors and organize the continuously recorded real-time passenger flow into a time series of historical passenger flow data;

[0080] S3: Calculate the passenger flow prediction value based on the feature vector of environmental information and the time series of historical passenger flow data; use the trained prediction model to calculate the passenger flow prediction value; extract the time series features in the time series of historical passenger flow data through the LSTM layer, and extract the non-time series features in the feature vector of environmental information through the fully connected layer;

[0081] S4: selecting a security strategy based on the passenger flow prediction value; presetting a passenger flow threshold, and selecting different security strategies according to the different magnitude relationships between the passenger flow prediction value and the passenger flow threshold;

[0082] S5: Set access control strategy, lighting control strategy and anomaly detection strategy based on the selected security strategy;

[0083] S6: Based on the set access control strategy, lighting control strategy and anomaly detection strategy, multi-scenario linkage security management and control of subway stations is carried out. Among them, the access control strategy controls the number of open entrances, exits and ticket inspection channels; the lighting control strategy controls the lighting brightness of each lighting device; and the anomaly detection strategy controls the detection cycle of anomaly detection based on surveillance video.

[0084] The specific functions of the above modules are realized by referring to the relevant contents of the subway site multi-scenario linkage safety management and control system described in Example 1 and will not be elaborated here.

[0085] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose and scope of protection of the present invention, which are all within the protection of the present invention.

Claims

1. The multi-scenario linkage safety control system for subway stations is characterized by: It includes access control module, monitoring module, lighting module, passenger flow prediction module, anomaly detection module and security strategy module; among which: The access control module is used to control the entry and exit of passengers and record the real-time passenger flow; The monitoring module is used to obtain the monitoring video of the subway station and provide the location of each monitoring video; The lighting module is used to control the working status of the lighting equipment in the subway station and monitor the status of each lighting equipment; The passenger flow prediction module calculates the passenger flow prediction value based on environmental information and historical passenger flow data; The anomaly detection module performs anomaly detection based on the surveillance video of the subway station and generates an abnormal alarm signal; The security strategy module selects a security strategy based on the passenger flow prediction value and the abnormal alarm signal; the access control module, the lighting module, and the abnormal detection module respectively adjust the access control strategy, the lighting control strategy, and the abnormal detection strategy based on the security strategy.

2. The subway station multi-scenario linkage safety management and control system according to claim 1, characterized in that: The access control module includes an access strategy unit and a counting unit; wherein the access strategy unit is configured with an access control strategy; the counting unit is used to record the passenger flow in the station in real time; The access control strategy includes a first access strategy and a second access strategy; wherein the first access strategy includes: the number of entrances open is , the open quantity for export is , the number of open ticket inspection channels is ; The first entry and exit strategy also includes The number of open entrances, The number of open exits, The number of the open ticket inspection channel; The second entry and exit strategy includes: the number of entrances opened is , the open quantity for export is , the number of open ticket inspection channels is ; The second entry and exit strategy also includes The number of open entrances, The number of open exits, The number of the open ticket inspection channel.

3. The subway station multi-scenario linkage safety management and control system according to claim 2, characterized in that: The monitoring module includes a camera unit and a map unit; wherein: the camera unit is used to obtain monitoring videos of the subway section from monitoring cameras distributed in the subway section; the map unit is used to provide an electronic map of the subway section; the electronic map provides the location of the monitoring camera corresponding to each monitoring video and the location of each sensor in the subway section.

4. The subway site multi-scenario linkage safety management and control system according to claim 3, characterized in that: The lighting module includes a monitoring unit and a control unit; wherein the monitoring unit is used to monitor the working state of each lighting device in real time; the control unit is configured with a lighting control strategy, and the lighting control strategy includes the lighting brightness of each lighting device in the subway section; the monitoring unit is configured with standard working parameters of each lighting device corresponding to the lighting control strategy, and if the working parameters of any lighting device are inconsistent with the corresponding standard working parameters, lighting abnormality information is sent to the map unit; the map unit marks the corresponding position of the lighting abnormality information in the electronic map; The lighting control strategy includes a first lighting strategy, a second lighting strategy, and a third lighting strategy; wherein, in the first lighting strategy, the lighting brightness of each lighting device in the subway section is the first brightness of the corresponding device; and the second lighting strategy includes the lighting brightness of each lighting device in the subway section being the second brightness of the corresponding device.

5. The subway station multi-scenario linkage safety management and control system as claimed in claim 4, characterized in that: The passenger flow prediction module includes a data collection unit and a prediction model unit; wherein the data collection unit is used to collect environmental information, obtain historical passenger flow data from the counting unit, and organize the historical passenger flow data into a time series; wherein the environmental information includes date information, time point, temperature, precipitation, and passenger flow of adjacent stations; wherein the date information includes holiday labels and week labels; the historical passenger flow data includes real-time passenger flow at the previous m time points, where m is a positive integer; The prediction model unit is configured with a prediction model, which is used to calculate the passenger flow prediction value; the input of the prediction model includes the feature vector of the environmental information and the time series of historical passenger flow data, and the output is the passenger flow prediction value within the next n minutes, where n is a positive integer.

6. The subway station multi-scenario linkage safety management and control system according to claim 5, characterized in that: The abnormality detection module includes a detection control unit; the detection control unit is configured with an abnormality detection strategy; the abnormality detection module also includes a first detection unit, which is used to perform grounding rod status detection, air conditioning cover detection, fire smoke detection, and switch jam detection through sensors; when any abnormal information of grounding rod abnormality, air conditioning cover not closed correctly, fire smoke, and switch jam is detected, the first detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the area coordinates corresponding to the abnormal alarm signal at the corresponding position in the electronic map.

7. The subway station multi-scenario linkage safety management and control system according to claim 6, characterized in that: The anomaly detection module also includes a second detection unit, which performs object left behind detection, animal intrusion detection, and train behavior detection based on the surveillance video of the subway station; wherein the method of performing object left behind detection and animal intrusion detection is as follows: collecting scene images that do not contain left behind objects and wild animals in the surveillance video, and establishing a scene background model; at fixed detection intervals, intercepting scene images from the surveillance video, and performing foreground detection to extract foreground images containing foreground objects; processing the foreground images through an image recognition model, and identifying and classifying foreground objects in the foreground images; The method of performing train behavior detection is as follows: at every fixed detection period, a monitoring image containing a train is intercepted from the monitoring video; a target detection algorithm is used to detect the train in the monitoring image; a multi-target tracking algorithm is used to continuously track the monitored train to obtain the train's motion trajectory, including the position and timestamp; the speed and dwell time of the train are calculated based on the train's motion trajectory; and train behavior detection is performed based on the speed and dwell time of the train; When any abnormal information such as abandoned items, wild animals, and abnormal train behavior is detected, the second detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the area coordinates corresponding to the abnormal alarm signal at the corresponding position in the electronic map.

8. The subway station multi-scenario linkage safety management and control system according to claim 7, characterized in that: The anomaly detection module further includes a third detection unit, which performs illegal intrusion detection based on the surveillance video of the subway station; the method is as follows: at fixed detection cycles, a surveillance image is captured from the surveillance video of each security control area, and a portrait detection is performed on the surveillance image; if a portrait is detected in the surveillance image, face recognition is performed on the portrait; If the face recognition result is an authorized manager, no illegal intrusion occurs; otherwise, illegal intrusion occurs; when illegal intrusion is detected, the third detection unit generates an abnormal alarm signal and sends the abnormal alarm signal to the security strategy module and the map unit respectively; the map unit marks the area coordinates corresponding to the abnormal alarm signal at the corresponding position in the electronic map.

9. The subway station multi-scenario linkage safety management and control system according to claim 8, characterized in that: The anomaly detection strategy includes a first detection strategy, a second detection strategy, and a third detection strategy; wherein the first detection strategy includes: the detection cycle of the second detection unit is , the detection cycle of the third detection unit is The second detection strategy includes: the detection cycle of the second detection unit is , the detection cycle of the third detection unit is .

10. The subway station multi-scenario linkage safety management and control system according to claim 9, characterized in that: The security strategy includes a first security strategy, a second security strategy, and a third security strategy; the security strategy module includes a strategy unit and an instruction unit; wherein the strategy unit is configured with a passenger flow threshold for selecting a security strategy; the strategy unit obtains a passenger flow prediction value from the prediction model unit, and if the passenger flow prediction value is not higher than the passenger flow threshold, the strategy unit selects the first security strategy; the instruction unit generates a first security instruction based on the first security strategy, and sends it to the entry and exit strategy unit, the control unit, and the detection control unit respectively; the entry and exit strategy unit responds to the first security instruction and triggers the first entry and exit strategy; the control unit responds to the first security instruction and triggers the first lighting strategy; the detection control unit responds to the first security instruction and triggers the first detection strategy; If the predicted passenger flow value is higher than the passenger flow threshold, the strategy unit selects the second security strategy; the instruction unit generates a second security instruction based on the second security strategy, and sends it to the entry and exit strategy unit, the control unit, and the detection control unit respectively; the entry and exit strategy unit responds to the second security instruction and triggers the second entry and exit strategy; the control unit responds to the second security instruction and triggers the second lighting strategy; the detection control unit responds to the second security instruction and triggers the second detection strategy.

11. The subway station multi-scenario linkage safety management and control system according to claim 10, characterized in that: If the strategy unit receives an abnormal alarm signal, the third security strategy is triggered; the instruction unit generates a third security instruction based on the third security strategy, and sends it to the control unit and the detection control unit respectively; the control unit responds to the third security instruction and triggers the third lighting strategy; the third lighting strategy is as follows: the coordinates of the area corresponding to the abnormal alarm signal are obtained through an electronic map, and the lighting brightness of each lighting device configured in the area corresponding to the abnormal alarm signal is set to a preset rated brightness; the detection control unit responds to the third security instruction and triggers the third detection strategy; the third detection strategy is as follows: the monitoring image of the corresponding area is collected at the rated brightness of the lighting device, and the abnormal detection is re-performed through the corresponding detection unit; if the corresponding detection unit sends an abnormal alarm signal, the abnormal alarm signal is pushed to the management personnel.

12. A multi-scenario linkage safety management and control method for a subway station, which is implemented based on the multi-scenario linkage safety management and control system for a subway station according to any one of claims 1 to 11, characterized in that: The following steps are involved: S1: Collect environmental information and continuously record real-time passenger flow; S2: Encode environmental information into feature vectors and organize the continuously recorded real-time passenger flow into a time series of historical passenger flow data; S3: Calculate the passenger flow forecast value based on the feature vector of environmental information and the time series of historical passenger flow data; S4: Selecting a security strategy based on the passenger flow prediction value; S5: Set access control strategy, lighting control strategy and anomaly detection strategy based on security strategy; S6: Perform multi-scenario linkage security management and control of subway stations based on the set access control strategy, lighting control strategy and anomaly detection strategy.

Images