WO2021159604A1 - Système de surveillance, procédé de surveillance et dispositif de surveillance pour train ferroviaire - Google Patents

Système de surveillance, procédé de surveillance et dispositif de surveillance pour train ferroviaire Download PDF

Info

Publication number
WO2021159604A1
WO2021159604A1 PCT/CN2020/085920 CN2020085920W WO2021159604A1 WO 2021159604 A1 WO2021159604 A1 WO 2021159604A1 CN 2020085920 W CN2020085920 W CN 2020085920W WO 2021159604 A1 WO2021159604 A1 WO 2021159604A1
Authority
WO
WIPO (PCT)
Prior art keywords
early warning
monitoring
preset
threshold
sound
Prior art date
Application number
PCT/CN2020/085920
Other languages
English (en)
Chinese (zh)
Inventor
崔蕾
孙少婧
吕白
韩璐
皮国瑞
戴忠贤
刘金海
Original Assignee
中车唐山机车车辆有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中车唐山机车车辆有限公司 filed Critical 中车唐山机车车辆有限公司
Priority to EP20919029.7A priority Critical patent/EP4105101A4/fr
Publication of WO2021159604A1 publication Critical patent/WO2021159604A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L15/00—Indicators provided on the vehicle or train for signalling purposes
    • B61L15/0072—On-board train data handling
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L15/00—Indicators provided on the vehicle or train for signalling purposes
    • B61L15/0081—On-board diagnosis or maintenance
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L23/00—Control, warning or like safety means along the route or between vehicles or trains
    • B61L23/04—Control, warning or like safety means along the route or between vehicles or trains for monitoring the mechanical state of the route
    • B61L23/041—Obstacle detection
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19608—Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
    • G08B13/19615—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion wherein said pattern is defined by the user
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19639—Details of the system layout
    • G08B13/19647—Systems specially adapted for intrusion detection in or around a vehicle

Definitions

  • This application relates to the technical field of rail trains, in particular to a monitoring system, monitoring method and monitoring device for rail trains.
  • the high-speed rail project carriage video monitoring system only has a passenger room camera to collect monitoring data, including video and audio, and store it in the video monitoring server to realize the driver's monitoring of the passenger room. It cannot analyze the data and warn of emergencies. .
  • the related track train monitoring system only collects monitoring data without analyzing the monitoring data, which is an urgent technical problem for those skilled in the art to solve.
  • the embodiments of the present application provide a monitoring system, a monitoring method, and a monitoring device for a rail train to solve the technical problem that the monitoring system of a related rail train only collects monitoring data without analyzing the monitoring data.
  • the embodiment of the present application provides a monitoring system for a rail train, including:
  • An acquiring device configured to acquire monitoring data in the rail train, where the monitoring data includes video
  • a monitoring server connected to the acquisition device to receive and store the monitoring data, and transmit the monitoring data to each analysis host;
  • a plurality of analysis hosts are respectively set in the carriages of the rail train, and are used to identify and analyze preset targets from the monitoring data. When the behavior of the preset targets meets the preset early warning conditions, report to the early warning device Send warning information;
  • An early warning device is connected to the monitoring server and is used to perform an early warning after receiving the early warning information forwarded by the monitoring server.
  • a monitoring method for rail trains includes the following steps:
  • a monitoring device for rail trains including:
  • An acquisition module configured to acquire monitoring data in the rail train, where the monitoring data includes video
  • the receiving and storing module is used to receive and store the monitoring data, and transmit the monitoring data;
  • the analysis module is used to identify and analyze a preset target from the monitoring data, and send early warning information when the behavior of the preset target meets a preset early warning condition;
  • the early warning module is used to receive early warning information and perform early warning.
  • the monitoring data in the rail train is acquired through the acquisition device, and the monitoring data is transmitted to each analysis host.
  • the analysis host is installed in the rail train. In this way, each analysis host analyzes a large amount of monitoring data separately, which can speed up the analysis;
  • the analysis host identifies and analyzes a preset target from the monitoring data, and when the behavior of the preset target meets a preset early warning condition, it sends early warning information to the early warning device, and the early warning device performs an early warning.
  • the monitoring system of the rail train in the embodiment of the application stores the monitoring data, which is convenient for later recalling and viewing, and also analyzes the monitoring data, and performs early warning when the preset early warning conditions are met, thereby realizing automatic analysis and early warning. It provides a basis for discovering special situations in time and implementing intervention measures as soon as possible.
  • FIG. 1 is a schematic structural diagram of a monitoring system for a rail train according to an embodiment of the application
  • FIG. 2 is a schematic diagram of the monitoring system shown in Figure 1 and installed on a rail train;
  • Figure 3 is a flow chart of the analysis host of the monitoring system shown in Figure 1 for analyzing whether there are dangerous behaviors;
  • Fig. 4 is a flow chart of analyzing whether the protected area is invaded by the analysis host of the monitoring system shown in Fig. 1;
  • Fig. 5 is a flow chart of analyzing whether people are crowded by the analysis host of the monitoring system shown in Fig. 1.
  • Fig. 1 is a schematic structural diagram of a monitoring system for a rail train according to an embodiment of the application
  • Fig. 2 is a schematic diagram of the monitoring system shown in Fig. 1 and installed on a rail train.
  • a monitoring system for a rail train includes:
  • the obtaining device 100 is configured to obtain monitoring data in the rail train, where the monitoring data includes video;
  • the monitoring server 200 is connected to the acquisition device to receive and store the monitoring data, and transmit the monitoring data to the analysis host;
  • a plurality of analysis hosts 300 are respectively arranged in the carriages of the rail train, and are used to identify and analyze a preset target from the monitoring data. When the behavior of the preset target meets a preset early warning condition, send an early warning The device sends early warning information;
  • the early warning device 400 is connected to the monitoring server and configured to perform an early warning after receiving the early warning information forwarded by the monitoring server.
  • the monitoring data in the rail train is obtained through the acquisition device, and the monitoring data is transmitted to each analysis host.
  • the analysis host is installed in the rail train. In this way, each analysis host monitors a large number of Data analysis can speed up the analysis; the analysis host recognizes and analyzes the preset target from the monitoring data, and when the behavior of the preset target meets the preset early warning conditions, the early warning information is sent to the early warning device to give an early warning.
  • the device carries out early warning.
  • the monitoring system of the rail train in the embodiment of the application stores the monitoring data, which is convenient for later recalling and viewing, and also analyzes the monitoring data, and performs early warning when the preset early warning conditions are met, thereby realizing automatic analysis and early warning. It provides a basis for discovering special situations in time and implementing intervention measures as soon as possible.
  • most rail trains are connected by Ethernet.
  • the acquisition device and the monitoring server are connected via Ethernet.
  • the Ethernet of the rail train is effectively used.
  • the amount of monitoring data is large and suitable for Data transmission via Ethernet.
  • the monitoring server and the analysis host are connected via Ethernet.
  • the analysis of the host computer uses the principle of time difference and the hybrid algorithm of background image difference, and uses the subtraction of adjacent frame images to extract the information of foreground target movement.
  • the process of video analysis background subtraction method: First, the system conducts background learning. The learning time varies according to the degree of background excitement, during which the system automatically establishes a background model. After that, the system enters the "analysis" state. If a moving object appears in the foreground and is within the set range and the preset target size meets the setting, the system will extract and track the preset target, and follow the pre-algorithm (intrusion, Remaining, fighting, etc.) trigger an early warning.
  • pre-algorithm intrusion, Remaining, fighting, etc.
  • the monitoring system Before triggering an early warning, the monitoring system has the function of identifying preset targets, which is to compare the extracted preset targets with the established model and select the best match.
  • the following describes how the monitoring system realizes early warning of dangerous behaviors.
  • Fig. 3 is a flow chart of analyzing whether there is a dangerous behavior by the analysis host of the monitoring system shown in Fig. 1.
  • the preset early warning conditions include reaching or exceeding the threshold of the action range;
  • the analysis host is specifically used to analyze whether there is a dangerous behavior, including step S300: identifying the preset target from the preset dangerous behavior monitoring range in the video of the rail train passenger compartment area and Analyze whether the action range of the preset target within the monitoring range of dangerous behavior reaches or exceeds the action range threshold:
  • Step S310 when reaching or exceeding the threshold value of the action amplitude, sending the early warning information of the dangerous behavior to the early warning device;
  • Step S320 When the action range of the preset target within the monitoring range of the dangerous behavior does not reach nor exceed the action range threshold, the warning information of the dangerous behavior is not sent;
  • the early warning device is specifically used for early warning of dangerous behaviors based on the early warning information of the dangerous behaviors.
  • an analysis of whether there is a dangerous behavior is performed, and the basis of the analysis is that the motion range threshold is reached or exceeded.
  • the analysis host determines that there is a dangerous behavior and gives an early warning;
  • the amplitude does not reach or exceed the threshold of the action amplitude, the analysis host determines that there is no dangerous behavior.
  • Dangerous behaviors can be set according to the situation, and various dangerous behaviors such as fights that require manual intervention by the staff of the rail train are included through the setting, so as to realize the early warning of dangerous behaviors.
  • the preset dangerous behavior monitoring range should not include the position where passengers need to carry out large-scale actions such as taking and unloading luggage, otherwise it is easy to cause misjudgment.
  • the range of an arbitrary polygon as the dangerous behavior monitoring range
  • the "audio detection mode”, "video detection mode” and “audio and video detection mode” three behavior detection Kind of.
  • the minimum detection size under CIF resolution is 64 ⁇ 32 pixels; the minimum response time is less than 2 seconds, and the detection success rate is greater than 80%.
  • the monitoring server sends the real-time video of the acquisition device in the passenger compartment area of the rail train to the analysis host by multicast, and the analysis host analyzes the video. Once someone has dangerous behavior, a corresponding warning will be generated.
  • the analysis of whether there is a dangerous behavior by the analysis host is achieved by obtaining a series of unique static and dynamic characteristics of the video image to realize the description and judgment of a specific event.
  • other computer vision and pattern recognition technologies such as optical flow, cluster analysis, image feature description, and classifiers are used.
  • the following describes how the monitoring system realizes the early warning of the intrusion of the protected area.
  • Fig. 4 is a flow chart of analyzing whether the protection area is invaded by the analysis host of the monitoring system shown in Fig. 1.
  • the preset early warning conditions also include preset targets in the protected area;
  • the analysis host is specifically used to analyze whether the protected area is invaded, and includes step S400: identifying whether there is a preset target from the preset intrusion behavior monitoring range in the video of the protected area:
  • Step S410 when the preset target exists, send the warning information that the protected area is invaded to the warning device;
  • Step S420 If the preset target does not exist, the early warning information that the protected area has been invaded is not sent to the early warning device;
  • the early warning device is specifically configured to perform an early warning of the intrusion of the protected area according to the early warning information of the intrusion of the protected area.
  • the early warning device is specifically configured to perform an early warning of the intrusion of the protected area according to the early warning information of the intrusion of the protected area.
  • the acquisition device deployed in the carriage can be deployed in key areas (such as the driver's cab, mechanic's room, etc.).
  • key areas such as the driver's cab, mechanic's room, etc.
  • the safety precautions for the key areas of the rail train can be realized And protection.
  • the basic method is to use the continuously input image sequence to obtain a background image as a reference.
  • the subsequent images are compared with the background image to obtain different pixels, and then these pixels are marked for connectivity. These marked areas are the initial Then, these targets are tracked to form a continuous tracking trajectory, and then the above-mentioned foreground and tracking trajectory are analyzed; compared with the preset rule information, the warning information is output.
  • Intrusion behavior monitoring range can be set to any polygonal intrusion behavior monitoring range
  • one or two of two behavior detections can be set: crossing the boundary of the intrusion behavior monitoring range and in the intrusion behavior monitoring range;
  • the direction of the crossing can be set to enter, leave or both directions. Whether the protection area is invaded or not is monitored under CIF resolution.
  • the minimum detection target size 10 ⁇ 10 pixels ,
  • the response time is less than 1 second, and the monitoring success rate is greater than 90%.
  • the following describes how the monitoring system realizes early warning of crowded people.
  • Fig. 5 is a flow chart of analyzing whether people are crowded by the analysis host of the monitoring system shown in Fig. 1.
  • the preset early warning conditions also include reaching or exceeding the threshold for the number of personnel;
  • the analysis host is specifically used to analyze whether people are crowded, including step S500: Whether the number of preset targets identified from the preset crowded monitoring range in the video of the designated area meets or exceeds the number of people Threshold:
  • Step S510 When the threshold of the number of personnel is reached or exceeded, the early warning information of crowded personnel is sent to the early warning device;
  • Step S520 When the threshold of the number of personnel has not been reached or exceeded, the early warning information of crowded personnel is not sent;
  • the early warning device is specifically configured to perform early warning of crowded people according to the early warning information of crowded people.
  • An acquisition device is set for locations where rail trains are prone to congestion, that is, the locations that are prone to congestion are designated areas, such as the aisle between the entrance of the rail train and the entrance of the passenger compartment; the congestion monitoring range preset in the video of the designated area, In the height range where the face is usually located, exclude the position where the face cannot reach at the height of the aisle.
  • the preset congestion monitoring range in the video of the designated area when the number of identified preset targets reaches or exceeds the threshold of the number of people, the early warning information of the people congestion is sent to the early warning device, so as to realize the early warning of the congestion.
  • the following describes how the monitoring system realizes the early warning of abnormal sound.
  • the monitoring data also includes audio;
  • the preset early warning conditions also include reaching or exceeding the voice early warning threshold;
  • the analysis host is also used to analyze whether the sound is abnormal, including: whether the sound level of the audio reaches or exceeds the sound warning threshold:
  • the sound warning threshold is not reached or exceeded, and the warning information of abnormal sound is not sent to the warning device;
  • the early warning device is also used to perform a sound abnormal warning based on the sound abnormal warning information.
  • the acquisition devices on both sides of the rail train are cameras with pickups, which can collect audio in real time, and the analysis host can analyze and process the audio. If the sound intensity exceeds the sound warning threshold, an abnormal sound warning is generated.
  • the intensity of the audio sound can be set; the shortest duration of the sound alarm can be set, the minimum response time for abnormal sound is 1 second, and the detection success rate is greater than 90%.
  • the monitoring system also includes a face database
  • the analysis host is also used to analyze whether key personnel on deployment are found, including: capturing face images from the video, and comparing the captured face images with the faces in the face database for recognition , When the captured face image matches the face in the face database, sending the early warning information of finding the key personnel to be deployed to the early warning device;
  • the early warning device is also used to perform an early warning for the discovery of key personnel based on the early warning information of the key personnel to be deployed and controlled.
  • the early warning of the key personnel needs to collect the information of the passengers, and through the analysis of the host comparison, the retrieval and early warning of the key personnel are realized.
  • the specific realization is that the acquisition device realizes the video capture, and the analysis host realizes the feature import of the database and the search and comparison query.
  • the analysis host is mainly divided into an image capture module, an image comparison module, a face database management module, and a face feature retrieval module:
  • Image capture module Capture real-time images in high-speed rail cars through the capture device in the rail train, and detect the face images in the captured video images every few frames and send them to the image Comparison module for comparison;
  • Image comparison module extract facial features in the image capture module
  • Face database management module manage the feature data of the face that has been recorded in the system
  • Face feature retrieval module retrieve whether the extracted face feature exists in the face database and determine whether it is the face of the key personnel.
  • the core of the early warning for key personnel is face recognition.
  • the face recognition algorithm includes three parts: face detection, face key detection, and face recognition. Face detection is to find all the faces contained in a picture, and face key point detection is to detect the key point coordinates of the face on the detected face image, and then estimate the posture of the face. Face recognition is Turn the face into a vector of a specific dimension, and judge whether it is a face image of the same person according to the similarity of the vector.
  • the analysis host performs at least one analysis of the monitoring data acquired by each acquisition device according to the setting position of the acquisition device, including analysis of whether there is a dangerous behavior, an analysis of whether a protected area is invaded, and personnel An analysis of whether it is crowded, an analysis of whether the sound is abnormal, and an analysis of whether a key deployment control personnel is found.
  • the monitoring data acquired by the acquiring device is analyzed separately, making full use of the monitoring data.
  • the protection area includes a driver's cab and/or a mechanic's room of a rail train.
  • the acquisition device includes a panoramic camera and a hemispherical camera with a sound pickup function
  • the panoramic camera 120 is set at the passing station of the rail train
  • the monitoring system also includes a monitoring screen, which is set in each room of the mechanic's room;
  • the monitoring screen is used to display real-time monitoring pictures and used to display the early warning information.
  • the preset target is identified and analyzed from the monitoring data, and when the behavior of the preset target meets the preset early warning condition, the step of sending early warning information specifically includes:
  • Analyzing whether there is a dangerous behavior including identifying and analyzing the preset target from the preset dangerous behavior monitoring range in the video of the rail train passenger compartment area, the preset target is within the monitoring range of the dangerous behavior Whether the action range of the behavior reaches or exceeds the action range threshold:
  • the warning message of dangerous behavior is sent; among them, the early warning conditions include reaching or exceeding the threshold of the action range;
  • the steps of receiving early warning information and performing early warning include:
  • the pre-warning of the dangerous behavior is carried out.
  • the preset target is identified from the monitoring data and analyzed, and when the behavior of the preset target reaches the preset early warning condition, the step of sending early warning information specifically includes:
  • Analyzing whether the protected area is invaded including identifying whether the preset target exists from the preset intrusion behavior monitoring range in the video of the protected area:
  • the preset target When the preset target exists, send an early warning message that the protected area is invaded; wherein, the early warning condition also includes the preset target in the protected area;
  • the steps of receiving early warning information and performing early warning include:
  • an early warning of the intrusion of the protected area is performed.
  • the preset target is identified from the monitoring data and analyzed, and when the behavior of the preset target reaches the preset early warning condition, the step of sending early warning information specifically includes:
  • the preset early warning conditions also include the threshold for the number of personnel being reached or exceeded;
  • the steps of receiving early warning information and performing early warning include:
  • an early warning of crowded people is performed.
  • the monitoring method of rail trains also includes:
  • the warning message of abnormal sound is sent; wherein the monitoring data also includes audio, and the preset warning conditions also include reaching or exceeding the sound warning threshold;
  • an abnormal sound warning is performed.
  • the monitoring method of rail trains also includes:
  • Analyze whether the key personnel to be deployed is found including: capturing face images from the video, and comparing the captured face images with the faces in the face database, and comparing the captured face images with When the faces in the face database are matched, sending an early warning message that key personnel are found to be deployed;
  • an early warning is carried out for the key personnel who are found to be deployed and controlled.
  • An acquisition module configured to acquire monitoring data in the rail train, where the monitoring data includes video
  • the receiving and storing module is used to receive and store the monitoring data, and transmit the monitoring data;
  • the analysis module is used to identify and analyze a preset target from the monitoring data, and send early warning information when the behavior of the preset target meets a preset early warning condition;
  • the early warning module is used to receive early warning information and perform early warning.
  • the analysis module includes:
  • the dangerous behavior analysis sub-module is used to analyze whether there are dangerous behaviors, including identifying and analyzing the preset target from the dangerous behavior monitoring range preset in the video of the rail train passenger compartment area, and in the dangerous behavior monitoring Whether the action range of the preset target within the range reaches or exceeds the action range threshold:
  • the warning message of dangerous behavior is sent; among them, the early warning conditions include reaching or exceeding the threshold of the action range;
  • the early warning module includes:
  • the dangerous behavior early warning sub-module is used for early warning of dangerous behaviors based on the warning information of the dangerous behaviors.
  • the analysis module further includes:
  • the intruded analysis sub-module is used to analyze whether the protected area is invaded, including: identifying whether the preset target exists from the preset intrusion behavior monitoring range in the video of the protected area:
  • the preset target exists, sending an early warning message that the protected area is invaded; wherein the preset early warning condition also includes a preset target in the protected area;
  • the early warning module also includes:
  • the intrusion early warning sub-module is used for the early warning information of the intrusion of the protected area, and the early warning of the intrusion of the protected area.
  • the analysis module further includes:
  • the crowding analysis sub-module is used to analyze whether people are crowded, including: whether the number of preset targets identified from the preset crowded monitoring range in the video of the designated area meets or exceeds the threshold of the number of people:
  • the preset early warning conditions also include the threshold for the number of personnel being reached or exceeded;
  • the early warning module also includes:
  • the congestion early warning sub-module is used to give early warning of crowdedness based on the early warning information of crowdedness.
  • the analysis module further includes:
  • the sound analysis sub-module is used to analyze whether the sound is abnormal, including whether the sound level of the audio reaches or exceeds the sound warning threshold:
  • the warning message of abnormal sound is sent; wherein the monitoring data also includes audio, and the preset warning conditions also include reaching or exceeding the sound warning threshold;
  • the early warning module also includes:
  • the sound abnormality early warning sub-module is used to perform sound abnormality early warning based on the sound abnormality warning information.
  • the analysis module further includes:
  • the key deployment control personnel analysis sub-module is used to analyze whether the key deployment control personnel are found, including: capturing face images from the video, and comparing the captured face images with the faces in the face database for recognition , When the captured face image matches the face in the face database, send an early warning message that the key personnel are found;
  • the early warning module also includes:
  • the early warning sub-module for the key personnel to be discovered is used to perform the early warning of the key personnel to be discovered based on the early warning information of the key personnel to be discovered.
  • connection can also be detachable or integrated; it can be mechanical, electrical, or communication; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication between two components. Or the interaction between two elements.
  • connection can also be detachable or integrated; it can be mechanical, electrical, or communication; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication between two components. Or the interaction between two elements.
  • the "on" or “under” of the first feature of the second feature may include direct contact between the first feature and the second feature, or include the first feature.
  • the second feature is not in direct contact but through another feature between them.
  • the "above”, “above” and “above” of the first feature on the second feature include the first feature directly above and obliquely above the second feature, or it simply means that the first feature is higher in level than the second feature.
  • the “below”, “below” and “below” of the first feature of the second feature include the first feature directly above and diagonally above the second feature, or it simply means that the level of the first feature is smaller than the second feature.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mechanical Engineering (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Train Traffic Observation, Control, And Security (AREA)
  • Alarm Systems (AREA)
  • Image Analysis (AREA)

Abstract

La présente invention concerne un système de surveillance pour train ferroviaire comprenant : des dispositifs d'acquisition (100) conçus pour acquérir des données de surveillance dans un train ferroviaire, les données de surveillance comprenant des vidéos ; un serveur de surveillance (200) connecté aux dispositifs d'acquisition (100) pour recevoir et stocker les données de surveillance et transmettre les données de surveillance à des ordinateurs principaux pour analyse (300) ; une pluralité d'ordinateurs principaux pour analyse (300) respectivement placés dans des wagons du train ferroviaire et configurés pour identifier une cible prédéfinie à partir des données de surveillance et pour l'analyser puis, lorsqu'un comportement de la cible prédéfinie satisfait une condition prédéfinie de pré-alarme, envoyer des informations de pré-alarme à un dispositif de pré-alarme (400) ; et le dispositif de pré-alarme (400) connecté aux ordinateurs principaux pour analyse (300) et configuré pour donner une pré-alarme lors de la réception des informations de pré-alarme. La présente invention concerne également un procédé de surveillance et un dispositif de surveillance. Le système de surveillance pour un train ferroviaire peut analyser des données de surveillance et déclencher une alarme à temps.
PCT/CN2020/085920 2020-02-12 2020-04-21 Système de surveillance, procédé de surveillance et dispositif de surveillance pour train ferroviaire WO2021159604A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP20919029.7A EP4105101A4 (fr) 2020-02-12 2020-04-21 Système de surveillance, procédé de surveillance et dispositif de surveillance pour train ferroviaire

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010089249.XA CN113256924A (zh) 2020-02-12 2020-02-12 一种轨道列车的监控系统，监控方法及监控装置
CN202010089249.X 2020-02-12

Publications (1)

Publication Number Publication Date
WO2021159604A1 true WO2021159604A1 (fr) 2021-08-19

Family

ID=77219738

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/085920 WO2021159604A1 (fr) 2020-02-12 2020-04-21 Système de surveillance, procédé de surveillance et dispositif de surveillance pour train ferroviaire

Country Status (3)

Country Link
EP (1) EP4105101A4 (fr)
CN (1) CN113256924A (fr)
WO (1) WO2021159604A1 (fr)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113850967A (zh) * 2021-10-18 2021-12-28 江苏立源自动化工程有限公司 基于煤炭装车系统的铁路货车运行安全与故障预警系统
CN114066157A (zh) * 2021-10-15 2022-02-18 江苏禹治流域管理技术研究院有限公司 一种采砂数据预警分析系统
CN114173220A (zh) * 2021-12-03 2022-03-11 中国铁道科学研究院集团有限公司标准计量研究所 铁路货物运输监控数据传输方法和装置
CN114863349A (zh) * 2022-06-30 2022-08-05 中关村科学城城市大脑股份有限公司 一种基坑安全防护监测方法及系统
CN115009324A (zh) * 2022-07-08 2022-09-06 成都西交轨道交通技术服务有限公司 基于云边协同的轨道交通不稳定网络事故预警系统及方法
CN115378968A (zh) * 2022-07-20 2022-11-22 慧之安信息技术股份有限公司 一种基于边缘计算的生物实验室安全检测系统
CN116363817A (zh) * 2023-02-02 2023-06-30 淮阴工学院 一种化工厂危险区域入侵预警方法及系统
CN116743970A (zh) * 2023-08-14 2023-09-12 安徽塔联智能科技有限责任公司 一种具有视频ai预警分析的智能管理平台
CN117251440A (zh) * 2023-09-20 2023-12-19 华安钢宝利高新汽车板加工(娄底)有限公司 一种行车电脑开发用数据传输自动纠正系统及方法
CN117395685A (zh) * 2023-10-30 2024-01-12 中国铁建电气化局集团有限公司 一种基于人工智能的高铁无线网络优化系统
CN117557966A (zh) * 2024-01-09 2024-02-13 南京格瑞物联科技有限公司 基于监控图像识别的校园异常行为安全检测方法及系统
CN118155357A (zh) * 2024-03-07 2024-06-07 广州市烨兴融集团有限公司 一种智慧办公的视频监控管理方法、系统、设备及介质
CN118296513A (zh) * 2024-06-04 2024-07-05 广东电网有限责任公司 变电站异物检测方法及系统

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114228794B (zh) * 2021-12-17 2023-09-22 神思电子技术股份有限公司 一种ctc调度的自动化监控方法及设备
CN114148378B (zh) * 2021-12-31 2022-07-19 安徽徽一通讯科技有限公司 铁路区间作业防护用声光报警器
CN114722102B (zh) * 2022-04-24 2023-01-13 中咨数据有限公司 一种基于大数据分析的轨道交通能耗系统智能监测管理系统
CN114715220A (zh) * 2022-06-07 2022-07-08 深圳龙铁高科技术有限公司 一种机车智能预警方法以及预警系统
CN116189305B (zh) * 2023-03-09 2023-07-18 合肥市轨道交通集团有限公司 一种基于神经网络模型嵌入的人员危险动作识别方法

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1710875A (zh) * 2005-06-07 2005-12-21 沈阳泽通数码有限公司 铁路列车数字化安全保障视频监控系统
EP2088566A2 (fr) * 2008-02-08 2009-08-12 Siemens Aktiengesellschaft Système de surveillance vidéo pour un véhicule et procédé de surveillance d'un véhicule
CN101795395A (zh) * 2009-02-04 2010-08-04 深圳市先进智能技术研究所 一种人群态势监控系统及方法
CN102254401A (zh) * 2011-05-11 2011-11-23 北京城市系统工程研究中心 一种客流运动智能分析方法
CN102385791A (zh) * 2011-11-08 2012-03-21 天津银箭科技有限公司 电子地图视频与报警监控联动系统
EP2945370A1 (fr) * 2013-01-10 2015-11-18 Mitsubishi Electric Corporation Système de surveillance d'intérieur de véhicule et procédé de surveillance d'intérieur de véhicule
CN205847433U (zh) * 2016-07-05 2016-12-28 北京清网华科技有限公司 一种车载视频监控系统
CN108297899A (zh) * 2018-02-07 2018-07-20 北京明铁科技有限公司 一种列车安全监控预警系统
CN108900814A (zh) * 2018-07-27 2018-11-27 安徽安为科技有限公司 一种列车远程视频监控系统
CN109543538A (zh) * 2018-10-23 2019-03-29 深圳壹账通智能科技有限公司 获取警情对象的轨迹的方法、装置、计算机设备和存储介质
CN110588729A (zh) * 2019-10-22 2019-12-20 西南交通大学 一种基于大数据的铁路监控辅助系统

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101174298A (zh) * 2006-11-02 2008-05-07 上海银晨智能识别科技有限公司 分散点大流量的人脸识别系统及识别方法
US20110316697A1 (en) * 2010-06-29 2011-12-29 General Electric Company System and method for monitoring an entity within an area
CN101917604A (zh) * 2010-09-03 2010-12-15 中铁上海设计院集团有限公司 用于轨道交通中的视频图像智能分析方法
CN202818505U (zh) * 2012-09-25 2013-03-20 毛振刚 列车车厢视频监控管理系统
CN103973653A (zh) * 2013-02-01 2014-08-06 上海迪爱斯通信设备有限公司 智能感知分析仪
CN103297751A (zh) * 2013-04-23 2013-09-11 四川天翼网络服务有限公司 智慧天网视频行为分析系统
US10392038B2 (en) * 2016-05-16 2019-08-27 Wi-Tronix, Llc Video content analysis system and method for transportation system
CN206100297U (zh) * 2016-08-16 2017-04-12 山西省交通科学研究院 一种列车车载监控系统
GB2556942A (en) * 2016-11-28 2018-06-13 Univ Of Lancaster Transport passenger monitoring systems
CN109606427B (zh) * 2019-01-11 2020-09-11 东北石油大学 基于ZigBee的列车车厢安全警告方法
CN110544360B (zh) * 2019-08-07 2022-04-08 北京全路通信信号研究设计院集团有限公司 一种列车安全驾驶监测系统及方法
CN110525456B (zh) * 2019-08-07 2020-11-24 北京全路通信信号研究设计院集团有限公司 一种列车安全驾驶监测系统及方法

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1710875A (zh) * 2005-06-07 2005-12-21 沈阳泽通数码有限公司 铁路列车数字化安全保障视频监控系统
EP2088566A2 (fr) * 2008-02-08 2009-08-12 Siemens Aktiengesellschaft Système de surveillance vidéo pour un véhicule et procédé de surveillance d'un véhicule
CN101795395A (zh) * 2009-02-04 2010-08-04 深圳市先进智能技术研究所 一种人群态势监控系统及方法
CN102254401A (zh) * 2011-05-11 2011-11-23 北京城市系统工程研究中心 一种客流运动智能分析方法
CN102385791A (zh) * 2011-11-08 2012-03-21 天津银箭科技有限公司 电子地图视频与报警监控联动系统
EP2945370A1 (fr) * 2013-01-10 2015-11-18 Mitsubishi Electric Corporation Système de surveillance d'intérieur de véhicule et procédé de surveillance d'intérieur de véhicule
CN205847433U (zh) * 2016-07-05 2016-12-28 北京清网华科技有限公司 一种车载视频监控系统
CN108297899A (zh) * 2018-02-07 2018-07-20 北京明铁科技有限公司 一种列车安全监控预警系统
CN108900814A (zh) * 2018-07-27 2018-11-27 安徽安为科技有限公司 一种列车远程视频监控系统
CN109543538A (zh) * 2018-10-23 2019-03-29 深圳壹账通智能科技有限公司 获取警情对象的轨迹的方法、装置、计算机设备和存储介质
CN110588729A (zh) * 2019-10-22 2019-12-20 西南交通大学 一种基于大数据的铁路监控辅助系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4105101A4 *

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114066157A (zh) * 2021-10-15 2022-02-18 江苏禹治流域管理技术研究院有限公司 一种采砂数据预警分析系统
CN113850967B (zh) * 2021-10-18 2022-05-27 江苏立源自动化工程有限公司 基于煤炭装车系统的铁路货车运行安全与故障预警系统
CN113850967A (zh) * 2021-10-18 2021-12-28 江苏立源自动化工程有限公司 基于煤炭装车系统的铁路货车运行安全与故障预警系统
CN114173220B (zh) * 2021-12-03 2023-06-27 中国铁道科学研究院集团有限公司标准计量研究所 铁路货物运输监控数据传输方法和装置
CN114173220A (zh) * 2021-12-03 2022-03-11 中国铁道科学研究院集团有限公司标准计量研究所 铁路货物运输监控数据传输方法和装置
CN114863349A (zh) * 2022-06-30 2022-08-05 中关村科学城城市大脑股份有限公司 一种基坑安全防护监测方法及系统
CN115009324B (zh) * 2022-07-08 2023-11-03 成都西交轨道交通技术服务有限公司 基于云边协同的轨道交通不稳定网络事故预警系统及方法
CN115009324A (zh) * 2022-07-08 2022-09-06 成都西交轨道交通技术服务有限公司 基于云边协同的轨道交通不稳定网络事故预警系统及方法
CN115378968B (zh) * 2022-07-20 2024-06-04 慧之安信息技术股份有限公司 一种基于边缘计算的生物实验室安全检测系统
CN115378968A (zh) * 2022-07-20 2022-11-22 慧之安信息技术股份有限公司 一种基于边缘计算的生物实验室安全检测系统
CN116363817B (zh) * 2023-02-02 2024-01-02 淮阴工学院 一种化工厂危险区域入侵预警方法及系统
CN116363817A (zh) * 2023-02-02 2023-06-30 淮阴工学院 一种化工厂危险区域入侵预警方法及系统
CN116743970B (zh) * 2023-08-14 2023-11-21 安徽塔联智能科技有限责任公司 一种具有视频ai预警分析的智能管理平台
CN116743970A (zh) * 2023-08-14 2023-09-12 安徽塔联智能科技有限责任公司 一种具有视频ai预警分析的智能管理平台
CN117251440A (zh) * 2023-09-20 2023-12-19 华安钢宝利高新汽车板加工(娄底)有限公司 一种行车电脑开发用数据传输自动纠正系统及方法
CN117251440B (zh) * 2023-09-20 2024-05-07 华安钢宝利高新汽车板加工(娄底)有限公司 一种行车电脑开发用数据传输自动纠正系统及方法
CN117395685A (zh) * 2023-10-30 2024-01-12 中国铁建电气化局集团有限公司 一种基于人工智能的高铁无线网络优化系统
CN117557966A (zh) * 2024-01-09 2024-02-13 南京格瑞物联科技有限公司 基于监控图像识别的校园异常行为安全检测方法及系统
CN117557966B (zh) * 2024-01-09 2024-04-02 南京格瑞物联科技有限公司 基于监控图像识别的校园异常行为安全检测方法及系统
CN118155357A (zh) * 2024-03-07 2024-06-07 广州市烨兴融集团有限公司 一种智慧办公的视频监控管理方法、系统、设备及介质
CN118296513A (zh) * 2024-06-04 2024-07-05 广东电网有限责任公司 变电站异物检测方法及系统

Also Published As

Publication number Publication date
EP4105101A1 (fr) 2022-12-21
EP4105101A4 (fr) 2023-08-02
CN113256924A (zh) 2021-08-13

Similar Documents

Publication Publication Date Title
WO2021159604A1 (fr) Système de surveillance, procédé de surveillance et dispositif de surveillance pour train ferroviaire
US10848719B2 (en) System and method for gate monitoring during departure or arrival of an autonomous vehicle
WO2019153193A1 (fr) Procédé et système de surveillance d'exploitation d'un taxi, dispositif et support de stockage
CN112132045B (zh) 一种基于计算机视觉的小区人员异常行为监控方案
JP6905850B2 (ja) 画像処理システム、撮像装置、学習モデル作成方法、情報処理装置
WO2019191142A1 (fr) Surveillance de zone intelligente avec une intelligence artificielle
KR102039277B1 (ko) 보행자 얼굴 인식 시스템 및 그 방법
WO2020024552A1 (fr) Procédé et système de surveillance de sécurité routière, et support de stockage lisible par ordinateur
KR101036947B1 (ko) 컴퓨터 영상 분석기술을 이용한 범죄 및 사고예방 자동경비 시스템
CN102081844A (zh) 一种交通视频行为分析与报警服务器
CN117749995B (zh) 一种基于多场景识别与语音交互的视频监控方法及系统
CN113676702B (zh) 基于视频流的目标追踪监测方法、系统、装置及存储介质
JP2004021495A (ja) 監視システムおよび監視方法
KR101459024B1 (ko) 시설물 감시용 보안 시스템
Gupta et al. CrowdVAS-Net: A deep-CNN based framework to detect abnormal crowd-motion behavior in videos for predicting crowd disaster
JP2008092279A (ja) 監視システム及び監視方法
KR20210043960A (ko) IoT와 인공지능 기술을 이용한 행동인식 기반의 안전 감시 시스템 및 방법
KR101407952B1 (ko) 엘리베이터 방범시스템 및 방법
Miller et al. Intelligent Sensor Information System For Public Transport–To Safely Go…
CN106572324A (zh) 一种节能型智能监控装置
CN111241918B (zh) 一种基于人脸识别的车用防跟踪方法及系统
JPH08180276A (ja) 障害物検知方法及びその装置
KR20100077662A (ko) 지능형 영상감시 시스템 및 영상감시 방법
WO2023124451A1 (fr) Procédé et appareil de génération d'événement d'alarme, dispositif et support de stockage
KR102407202B1 (ko) 지능형 영상 분석 장치 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20919029

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020919029

Country of ref document: EP

Effective date: 20220912