WO2020155294A1 - 一种基于机器视觉技术的地铁站台门障碍物探测预警系统 - Google Patents
一种基于机器视觉技术的地铁站台门障碍物探测预警系统 Download PDFInfo
- Publication number
- WO2020155294A1 WO2020155294A1 PCT/CN2019/077231 CN2019077231W WO2020155294A1 WO 2020155294 A1 WO2020155294 A1 WO 2020155294A1 CN 2019077231 W CN2019077231 W CN 2019077231W WO 2020155294 A1 WO2020155294 A1 WO 2020155294A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- camera
- platform door
- door
- train
- subway platform
- Prior art date
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61B—RAILWAY SYSTEMS; EQUIPMENT THEREFOR NOT OTHERWISE PROVIDED FOR
- B61B1/00—General arrangement of stations, platforms, or sidings; Railway networks; Rail vehicle marshalling systems
- B61B1/02—General arrangement of stations and platforms including protection devices for the passengers
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
Definitions
- the invention relates to the technical field of monitoring and early warning, in particular to a subway station door obstacle detection and early warning system based on machine vision technology.
- the above-mentioned systems are all passive detection methods.
- the detection action only occurs after the subway platform door system and the train signal are handed over.
- the above detection methods cannot identify the types of obstacles, let alone detect the dangerous behaviors of the obstacles. Therefore, the detection range of the passive detection system is far from meeting the current operating requirements of the subway platform door system.
- the embodiment of the present invention provides a subway station door obstacle detection and early warning system based on machine vision technology, and aims to provide a detection and early warning system with higher timeliness, reliability and safety.
- a subway station door obstacle detection and early warning system based on machine vision technology includes: a camera, a buffer device communicatively connected with the camera, a monitoring unit communicatively connected with the buffer device and the camera, and an alarm device communicatively connected with the monitoring unit;
- two mechanical structures are respectively arranged on both sides of the subway platform door, and two cameras are respectively installed on the two mechanical structures;
- the mechanical structure includes: a camera mounting bracket fixed on the ground, and an adjustable mounting bracket The camera installation adjustment mechanism at the upper end of the camera installation bracket is provided with a camera installation position.
- the mechanical structure further includes a camera outer protective cover and a transparent protective plate installed on the upper end of the camera mounting bracket for protecting the camera.
- the camera outer protective cover accommodates the entire camera in its internal protective space.
- the guard plate is located in front of the camera lens.
- the subway platform door includes a fixed door unit on both sides and a sliding door unit in the middle, the two cameras are respectively arranged on the outer side of the fixed door unit on both sides, and the detection area refers to the sliding door.
- the camera is used to collect images of obstacles in the detection area; the buffer device is used to store the collected image data in time series; the monitoring unit is used to analyze and process the collected image data and identify obstacles Type, judge whether there is danger, and issue an alarm instruction when the danger is recognized; the alarm device is used to alarm when the alarm instruction is received.
- the monitoring unit is also used for classifying and sorting the image data from the cache device, and the sorting and sorting includes: sorting by time sequence, sorting by obstacle category, and sorting by risk.
- the monitoring unit is also used to communicate with the subway platform door control unit and provide monitoring data, so that the subway platform door control unit combines the monitoring data obtained to control the subway platform door.
- the camera is also communicatively connected with the central monitoring center of the station and provides monitoring data.
- the camera, the buffer device, the monitoring unit, the alarm device, and the station central monitoring center jointly constitute an obstacle detection and recognition system; the obstacle detection and recognition system is used for train entry Obstacle recognition in front of the station or before leaving the station. If there is an obstacle, the control train cannot enter or leave the station; and, it is used to identify the obstacle in the train stop, judge whether there is a danger near the platform door, and execute according to the judgment result Operate accordingly.
- the alarm device is also used to issue a train entry warning, a train stop warning, and a train exit warning.
- the subway station door obstacle detection and early warning system based on the machine vision technology of the present invention uses machine vision technology and has three-dimensional attributes. It can judge the occurrence time of the detected object obstacle in real time, the type and the location of the occurrence and other key information, which can be integrated Evaluate whether the obstacle is dangerous, and remove the dangerous obstacle or its behavior before the platform door and the train signal handover. Its timeliness, reliability and safety are better than traditional detection and early warning systems.
- FIG. 1 is a block diagram of the principle block diagram of a subway platform door obstacle detection and early warning system based on machine vision technology according to an embodiment of the present invention
- FIG. 2 is a schematic structural diagram of a subway platform door obstacle detection and early warning system based on machine vision technology in an embodiment of the present invention
- FIG. 3 is a schematic diagram of a detection area of a subway platform door obstacle detection and early warning system based on machine vision technology in an embodiment of the present invention.
- an embodiment of the present invention provides a subway station door obstacle detection and early warning system based on machine vision technology (herein referred to as the detection and early warning system).
- the detection and early warning system includes:
- the two cameras 1 are respectively installed on the mechanical structure provided on both sides of the subway platform door, and are at a certain height from the ground, and face the detected area of the subway platform door.
- the subway platform door includes a fixed door unit located on both sides and a sliding door unit located in the middle, and the two cameras 1 are respectively arranged outside the fixed door units on both sides.
- the detection area refers to the passage area formed by the sliding door unit and the gap between the fixed door unit and the train. As shown in Figure 3, the present invention divides the detection area 9 into left and right defense zones according to the left and right.
- two mechanical structures are respectively installed on both sides of the subway station door, and two cameras 1 are installed on the two mechanical structures respectively; the two cameras 1 work together and are arranged facing each other to perform video capture on the detected area.
- the detection area forms a camera cross-overlap shot without blind spots and a large area shot.
- the shooting area can reach 1600mm (height) * 2500mm (width), which can complete a large area without dead ends and high-precision detection.
- the camera is usually installed at a height of 1.2 meters or above.
- the two cameras 1 are arranged on the outside of the fixed door units on both sides through two mechanical structures, which can expand the detection range of obstacles and reduce the visual impact on passengers. At the same time, they are far away from the sliding door unit to avoid emergency evacuation. The blocking effect of the time.
- the mechanical structure may include: a camera mounting bracket 2 fixed on the ground, and an adjustable camera mounting adjustment mechanism 3 mounted on the upper end of the camera mounting bracket 2.
- the camera mounting adjustment mechanism 3 is provided with a camera installation position for easy installation Camera 1.
- the mechanical structure may further include a camera outer protective cover 4 and a transparent protective plate 5 installed on the upper end of the camera mounting bracket 2 to protect the camera 1, and the camera outer protective cover 4 accommodates the entire camera inside.
- the transparent protective plate 5 is located in front of the camera lens.
- the transparent shield has the function of isolating the camera and related equipment from the platform environment, preventing the dust caused by subway operation from damaging the lens, and also protecting the equipment from deliberate or unintentional damage by passengers.
- the built-in camera design can make the system withstand the impact of dust in the harsh environment of subway operation.
- the camera mounting bracket 2 can be installed on the ground platform door platform plate through the equipment fixing platform to realize the stability design and the vibration isolation design.
- the equipment fixing stand can be connected to the camera mounting bracket 2 by welding to avoid the movement of the installation position due to the weight and vibration of the camera mounting bracket 2.
- an arc-shaped long hole can be provided on the equipment fixing stand, which forms an adjusting mechanism with bolts, so that the camera mounting bracket 2 can be installed in a better position.
- the camera installation adjustment mechanism 3 is provided with an overhaul hole for installing fine adjustment and mechanical locking equipment to have installation adjustability and mechanical locking performance.
- the mechanical structure has installation adjustability, can adapt to the special installation requirements of the rail transit operating environment, and the outer protective cover and the transparent protective plate make the physical system have high waterproof and dustproof performance and prevent damage to external objects.
- the camera 1, the cache device 6, the monitoring unit 7, and the alarm device 8 together constitute the monitoring system.
- the camera mounting bracket 2, the camera mounting adjustment mechanism 3, the camera outer protective cover 4 and the transparent shield 5 together constitute the mechanical structure, the monitoring system and The mechanical structure together constitute the detection and early warning system of the present invention.
- the camera 1 is used to collect images of obstacles in the detection area, especially real-time images of dangerous obstacles and their behaviors, and quickly transmit the collected image data to the cache device 6 and the monitoring unit 7 according to specific envoys. ;
- the buffer device 6 is used to store the collected image data in time series, especially to store real-time images of dangerous obstacles and their behaviors in time series; it is convenient for the monitoring unit to quickly download and use relevant image data;
- the monitoring unit 7 can perform active obstacle recognition, which is specifically used to analyze and process the collected image data, identify the obstacle category, and give feedback information according to the determination standard, that is, determine whether there is danger, for example, determine whether there are passengers Trapped in the gap between the fixed door unit and the train, and issue an alarm command when a danger is identified;
- the alarm device 8 is used to give an alarm when an alarm instruction is received.
- the alarm device 8 can be an independent alarm device installed near the subway platform door, or it can use the existing platform door early warning system, or both Both.
- the monitoring unit 7 can determine whether the platform door early warning system will give an alarm according to the condition of the obstacle, or an independent alarm device will give an alarm.
- the monitoring unit 7 is also used to identify and sort the image data from the cache device.
- the sorting and sorting includes: sorting by time sequence, sorting by obstacle category, and sorting by risk.
- the image data obtained by the camera is mainly embodied in the time sequence of recording the image, the type of the image content, the main body of the image content and the location where the actions occur, etc.
- the monitoring unit 7 is also used to communicate with the subway platform door control unit and provide monitoring data, so that the subway platform door control unit combines the monitoring data obtained to control the subway platform door.
- the monitoring unit 7 can actively input real-time images to the subway platform door control unit, or, according to the requirements of the subway platform door control unit, retrieve and transmit image data of any past time period from the cache device.
- the camera 1 is also communicatively connected with the central monitoring center of the station and provides monitoring data.
- the image data can be viewed remotely online, and the image data can be further viewed wirelessly through the client terminal.
- Online real-time manual access can be performed through the station central monitoring center, and further online real-time manual access can be performed by authorized clients through the station central monitoring center.
- the monitoring unit 7 can directly monitor the camera 1 and can read data online in real time when needed. At the same time, the working status of the camera 1 is controlled by the monitoring unit 7 in real time.
- the monitoring unit 7 reports information to the central monitoring center of the station, and the central monitoring center of the station can read the image data of the camera 1 online in real time when needed, so that the central monitoring center of the station can check and judge the obstacle category and whether manual intervention is required.
- the camera 1, the buffer device 6, the monitoring unit 7, the alarm device 8, and the station central monitoring center together constitute an obstacle detection and recognition system.
- the obstacle detection and recognition system is used to identify obstacles before the train enters or exits the station. If there is an obstacle, the train can not enter or exit the station; and, it is used to identify the obstacle when the train is stopped. , Judge whether there is a danger near the platform door, and perform corresponding operations according to the judgment result.
- the obstacle detection and scanning can be carried out in the defense area, and the image data is recorded by the camera.
- the monitoring unit compares the image data input by the camera. If there is an obstacle, the train cannot Inbound or outbound. When the train is stopping, passengers get on and off the train through the sliding door passage. During this period, obstacles may enter the gap between the train and the platform intentionally or unintentionally.
- the camera can take real-time photography and the monitoring unit can identify possible dangers. And transmit the judgment result to the subway platform door control unit.
- the alarm device 8 is also used to issue a train entry warning, a train stop warning, and a train exit warning.
- the camera can be used to record images of obstacles in the station defense zone (that is, in the detection area) in real time, and input to the buffer device and monitoring unit.
- the monitoring unit can be used to classify the image information, and give feedback information according to the judgment standard, that is, judge whether it is a dangerous obstacle, if it is, send an alarm instruction, if it is not, track until the danger disappears.
- the early warning system can be used for real-time alarm in the application scenario of the platform door, and at the same time, the judgment result can be sent to the central monitoring center of the station, which is convenient to determine whether the authorized personnel of the station perform manual intervention.
- the central monitoring center of the station can determine the information fed back by the monitoring unit Whether to perform authorized manual intervention.
- classification process described in this article refers to the identification and classification of obstacles entering the detection area.
- classification according to time sequence or time classification may include: before the train enters the station, during which the train stops and gets on and off passengers through the sliding door passage, and before the train leaves the station, etc.
- Classification by obstacle category or image content can include: passengers, small animals, objects, items carried by passengers (image content), and dangerous behaviors of passengers or objects (behavior of the image subject).
- the camera described can be an industrial camera, such as an industrial-grade color camera.
- the color camera has the characteristics of small size, low power, and POE power supply. Therefore, the color camera is a best practice case of the present invention. Infrared cameras and other types of cameras It can also be applied to the system and should fall within the protection scope of the present invention.
- the detection area is also called the detection alarm zone.
- the detection alarm zone in this article is expanded from the traditional sliding door channel area to the fixed door adjacent to the sliding door channel.
- the unit includes the gap between the fixed door unit and the train.
- the red alarm of the present invention has changed from passive to active.
- the active is mainly reflected in the monitoring unit identifying and judging dangerous behaviors, such as passengers crossing the platform door on the platform side and throwing to the track side
- the monitoring unit will immediately issue an alarm instruction, and the central monitoring system of the station can determine whether it is necessary to authorize personnel to manually intervene to broadcast warnings or notify station attendants to stop manually according to the actual situation.
- the invention moves the detection and alarm of obstacles before the train enters or exits the station.
- the detection and early warning system of the present invention can be used on the half-height security door system of subway platform doors, and can also be used on the screen door system.
- the embodiments of the present invention provide a subway platform door obstacle detection and early warning system based on machine vision technology.
- the embodiments of the present invention have the following advantages:
- the subway station door obstacle detection and early warning system based on the machine vision technology of the present invention uses machine vision technology and has three-dimensional attributes. It can judge the occurrence time of the detected object obstacle in real time, the type and the location of the occurrence and other key information, which can be integrated Evaluate whether the obstacle is dangerous, and remove the dangerous obstacle or its behavior before the platform door and the train signal handover. Its timeliness, reliability and safety are better than traditional detection and early warning systems.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
一种基于机器视觉技术的地铁站台门障碍物探测预警系统,包括摄像机(1)、与摄像机通信连接的缓存设备(6)、与缓存设备及摄像机通信连接的监控单元（7）、以及与监控单元通信连接的报警装置（8），其中两台摄像机安装在地铁站台门的两侧，距离地面一定高度，并朝向地铁站台门的被探测区域。该探测预警系统能实施判断对象障碍物的发生时间、所属种类和发生位置，提高探测预警系统的时效性、可靠性和安全性。
Description
本发明涉及监控预警技术领域，具体涉及一种基于机器视觉技术的地铁站台门障碍物探测预警系统。
考虑车辆运行时的晃动以及地铁站台门受各种载荷产生变形的影响，在地铁站台门与列车之间预留有一定的间隙。然而，正是该间隙的存在，为地铁系统运行引入了新的安全隐患。特别是有乘客被夹伤或被困在此间隙中，对地铁运行系统造成的损失和影响将十分严重。因此设置稳定、可靠的障碍物探测预警系统显得尤为重要。
目前有红外线或激光等探测预警系统，但上述系统均为被动探测方式，探测动作只发生在地铁站台门系统和列车信号交接之后，存在时效性低、可靠性差和误报率高的缺点。同时上述探测方式不能对障碍物的种类进行识别，更不能对障碍物所具有的危险性行为进行探测，因此被动式探测系统的探测范围远远不能满足目前地铁站台门系统的运行需求。
本发明实施例提供一种基于机器视觉技术的地铁站台门障碍物探测预警系统，旨在提供一种时效性、可靠性和安全性更高的探测预警系统。
本发明采用的技术方案如下：
一种基于机器视觉技术的地铁站台门障碍物探测预警系统，包括：摄像机，与摄像机通信连接的缓存设备，与缓存设备及摄像机通信连接的监控单元，以及与监控单元通信连接的报警装置；所述摄像机有两台，分别安装在地铁站台门两侧所设的机械结构上，且距离地面一定高度，并朝向地铁站台门的被探测区域。
一种实现方式中，两部机械结构分别设置在地铁站台门两侧，两台摄像机分别安装在两部机械结构上；机械结构包括：固定在地面上的摄像机安装支架，和可调节的安装在摄像机安装支架上端的摄像机安装调节机构，所述摄像机安装调节机构上设有摄像机安装位。
进一步的，所述机械结构还包括安装在摄像机安装支架上端、用于保护摄像机的摄像机外防护罩和透明护板，所述摄像机外防护罩将整个摄像机容纳在其内部保护空间内，所述透明护板位于摄像机镜头前。
一种实现方式中，所述地铁站台门包括位于两边的固定门单元和位于中间的滑动门单元，所述两台摄像机分别设置在两边的固定门单元的外侧，所述探测区域是指滑动门单元所形成的通道区域以及固定门单元与列车之间的间隙。
一种实现方式中，所述摄像机用于采集探测区域内障碍物的影像；所述缓存设备用于按时序储存采集的影像数据；所述监控单元用于分析处理采集的影像数据，识别障碍物类别，判断是否有危险，并在识别出危险时发出报警指令；所述报警装置用于在收到报警指令时进行报警。
进一步的，所述监控单元，还用于对来自缓存设备的影像数据进行分类整理，所述分类整理包括：按时序分类、按障碍物类别分类和按照危险性分类。
进一步的，所述监控单元还用于与地铁站台门控制单元通信连接并提供监控数据，以使地铁站台门控制单元结合得到的监控数据对地铁站台门实施控制。
进一步的，所述摄像机还与车站中央监控中心通信连接并提供监控数据。
进一步的，所述摄像机、所述缓存设备、所述监控单元和所述报警设备以及所述车站中央监控中心，共同构成障碍物探测识别系统；所述障碍物探测识别系统，用于进行列车进站前或出站前障碍物识别，如果存在障碍物，控制列车不能进站或出站；以及，用于进行列车停靠中障碍物识别，对站台门附近是否存在危险进行判断，根据判断结果执行相应操作。
进一步的，所述报警装置还用于发出列车进站预警、列车停靠预警和列车出站预警。
从以上技术方案可以看出，本发明实施例具有以下优点：
本发明的基于机器视觉技术的地铁站台门障碍物探测预警系统，使用机器视觉技术，具有三维属性，能够实时判断探测对象障碍物发生的时间，所属的种类和发生的位置等关键信息，能够综合评价障碍物是否具有危险性，并将具有危险性的障碍物或其行为解除在站台门和列车信号交接之前，其时效性、可靠性和安全性优于传统的探测预警系统。
为了更清楚地说明本发明实施例技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1是本发明实施例基于机器视觉技术的地铁站台门障碍物探测预警系统的原理框图；
图2是本发明实施例基于机器视觉技术的地铁站台门障碍物探测预警系统的结构示意图；
图3是本发明实施例基于机器视觉技术的地铁站台门障碍物探测预警系统的探测区域的示意图。
为了使本技术领域的人员更好地理解本发明方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分的实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语 “包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。
下面通过具体实施例，分别进行详细的说明。
请参考图1和图2，本发明实施例提供一种基于机器视觉技术的地铁站台门障碍物探测预警系统（本文简称为探测预警系统）。该探测预警系统包括：
摄像机1，与摄像机1通信连接的缓存设备6，与缓存设备6及摄像机1通信连接的监控单元7，以及与监控单元7通信连接的报警装置8。所述两台摄像机1分别安装在地铁站台门两侧所设的机械结构上，且距离地面一定高度，并朝向地铁站台门的被探测区域。
其中，所述地铁站台门包括位于两边的固定门单元和位于中间的滑动门单元，所述两台摄像机1分别设置在两边的固定门单元的外侧。所述探测区域是指滑动门单元所形成的通道区域以及固定门单元与列车之间的间隙。如图3所示，本发明将所述探测区域9按照左右分为左防区和右防区。
其中，两部机械结构分别设置在地铁站台门两侧，两台摄像机1分别安装在两部机械结构上；所述两台摄像机1协同工作，朝向相对设置，对被探测区域进行视频采集，在探测区域形成摄像交叉重叠无死角拍摄和大区域拍摄。通过多摄像机协同采集，拍摄区域可达1600mm（高）*2500mm（宽），完成大区域，无死角，高精度探测。
优选的，摄像机通常安装在1.2米或以上高度的位置。
两台摄像机1通过两部机械结构分别设置在两边的固定门单元的外侧，可起到扩大障碍物探测范围的作用，并降低对乘客的视觉影响，同时远离滑动门单元，避免因在紧急疏散时产生的阻滞影响。
机械结构可包括：固定在地面上的摄像机安装支架2，和可调节的安装在摄像机安装支架2上端的摄像机安装调节机构3，所述摄像机安装调节机构3上设有摄像机安装位，以便于安装摄像机1。
可选的，所述机械结构还可包括安装在摄像机安装支架2上端、用于保护摄像机1的摄像机外防护罩4和透明护板5，所述摄像机外防护罩4将整个摄像机容纳在其内部保护空间内，所述透明护板5位于摄像机镜头前。透明护板具有使摄像机及相关设备与站台环境隔离的作用，防止地铁运行引起的粉尘损坏镜头，同时也可以保护设备免受乘客故意或无意的破坏。也就是说，摄像机内嵌式设计可使系统耐受地铁运行恶劣环境粉尘的影响。
可选的，摄像机安装支架2可通过设备固定台架安装在地面站台门站台板上，实现稳定性设计和隔振设计。设备固定台架可通过焊接方式与摄像机安装支架2连接，避免因摄像机安装支架2重量和振动而产生安装位置的移动。同时设备固定台架上可设有弧形长孔，与螺栓构成调节机构，从而使摄像机安装支架2安装在较佳位置。
可选的，摄像机安装调节机构3，设置有检修孔，用于安装微调节和机械锁紧设备，以具有安装可调节性和机械锁紧性能。
所述机械结构具有安装可调节性，能够适应轨道交通运行环境的特殊安装需求，以及外防护罩和透明护板使该物理系统具有高防水、防尘性能以及防止外物损坏的性能。
如上，摄像机1、缓存设备6、监控单元7、报警装置8共同构成监控系统，摄像机安装支架2、摄像机安装调节机构3、摄像机外防护罩4和透明护板5共同构成机械结构，监控系统和机械结构一起构成本发明的探测预警系统。
下面，对监控系统各个组成单元的功能说明如下：
所述摄像机1，用于采集探测区域内障碍物的影像，尤其是具有危险性的障碍物及其行为的实时影像，并将采集的影像数据按特定特使快速传输至缓存设备6及监控单元7；
所述缓存设备6，用于按时序储存采集的影像数据，尤其是按时序对具有危险性的障碍物及其行为的实时影像进行储存；便于监控单元快速下载使用相关影像数据；
所述监控单元7，可进行主动式障碍物识别，具体用于分析处理采集的影像数据，识别障碍物类别，根据判定标准给出反馈信息，即，判断是否有危险，例如判断是否有乘客被困入固定门单元与列车之间的间隙，并在识别出危险时发出报警指令；
所述报警装8，用于在收到报警指令时进行报警，该报警装置8可以是安装在地铁站台门附近的独立报警装置，或者，也可以采用现有的站台门预警系统，或者同时采用两者。
监控单元7可根据障碍物的情况决定是由站台门预警系统进行报警，或是由独立报警装置进行报警。
进一步的，所述监控单元7，还用于对来自缓存设备的影像数据进行识别和分类整理，所述分类整理包括：按时序分类、按障碍物类别分类和按照危险性分类。所述摄像机所获取的影像数据，主要体现在记录影像的时序、影像内容的种类和影像内容主体及其行为所发生的位置等。
进一步的，所述监控单元7，还用于与地铁站台门控制单元通信连接并提供监控数据，以使地铁站台门控制单元结合得到的监控数据对地铁站台门实施控制。所述监控单元7，可以主动向地铁站台门控制单元输入实时影像，或者，也可以根据地铁站台门控制单元的需求，从缓存设备中调取并传输任意过往时段的影像数据资料。
进一步的，所述摄像机1还与车站中央监控中心通信连接并提供监控数据。通过车站中央监控中心可进行远程在线察看影像数据，以及可进一步通过客户端远程无线察看影像数据。通过车站中央监控中心可进行在线实时人工接入，以及可进一步由授权的客户端通过车站中央监控中心进行在线实时人工接入。
监控单元7可以直接对摄像机1进行监控，需要时可以实时在线读取数据，同时摄像机1的工作状态受监控单元7实时控制。监控单元7向车站中央监控中心进行信息汇报，同时车站中央监控中心需要时可以实时在线读取摄像机1的影像数据，以便车站中央监控中心可以查看和判断障碍物类别和是否需要人工介入。
进一步的，所述摄像机1、所述缓存设备6、所述监控单元7和所述报警设备8以及所述车站中央监控中心，共同构成障碍物探测识别系统。所述障碍物探测识别系统，用于在列车进站前或出站前进行障碍物识别，如果存在障碍物，控制列车不能进站或出站；以及，用于在列车停靠中进行障碍物识别，对站台门附近是否存在危险进行判断，根据判断结果执行相应操作。
具体的：在列车进站前或出站前，可对防区内进行障碍物探测扫描，并通过摄像机记录影像数据，监控单元将摄像机所输入的影像数据进行比较对照，如果存在障碍物，列车不能进站或出站。在列车停靠中，乘客通过滑动门通道上下车，值此期间，可能会有障碍物故意或无意间进入列车与站台之间缝隙，摄像机可实时摄影，并由监控单元对可能的危险进行识别，并将判定结果传至地铁站台门控制单元。
进一步的，所述报警装置8，还用于发出列车进站预警、列车停靠预警和列车出站预警。
如上所述，所述摄像机可用于实时摄录站台防区内（即探测区域内）的障碍物影像，并输入到所述的缓存设备和监控单元。所述监控单元可用于将影像信息进行分类处理，并根据判定标准给出反馈信息，即判断是否为具有危险性的障碍物，如果是，则发出报警指令，如果不是，则跟踪直至危险消失。所述预警系统可用于在站台门应用场景下实时报警，同时可将判定结果输送至车站中央监控中心，便于确定车站授权人员是否进行人工介入，车站中央监控中心可结合监控单元所反馈的信息决定是否进行具有授权的人工介入。
值得说明的是，本文所述的分类处理是指对进入探测区域内的障碍物进行识别和分类。其中，所述的按时序分类或者说按时间分类可包括：列车进站前，列车停稳并通过滑动门通道上下乘客期间，列车出站前等不同时段。按障碍物类别或者说按影像内容分类可包括：乘客、小动物、物体，乘客随身携带物品（影像内容），乘客或物体的危险行为（影像主体的行为）。
本文中，所述的摄像机可以采用工业摄像机，例如工业级彩色摄像机，彩色摄像机具有体积小，功率低，POE供电等特点，因此彩色摄像机是本发明的较佳实践案例，红外摄像机以及其他类型摄像机同样可以应用到该系统中，应属本发明的保护范围。
本文中，所述的探测区域，也称为探测报警防区，相比于传统探测预警系统的防区，本文中的探测报警防区由传统的滑动门通道区域，扩大至滑动门通道相邻的固定门单元，包括固定门单元与列车之间的间隙。
相比于传统的预测预警系统，本发明红报警由被动式变为主动式，主动式主要体现在监控单元对具有危险的行为进行识别和判断，例如乘客在站台侧越过站台门，向轨道侧抛掷物体，或强行进入防区，当上述行为发生时，监控单元立即发出报警指令，并可由车站中央监控系统根据实际情况，判断是否需要授权人员进行人工介入进行广播警告或通知站务人员进行人工阻止。本发明将障碍物的探测和报警前移至列车进站前或出站前。
需要说明的是，本发明的探测预警系统可以用在地铁站台门的半高安全门系统上，也可以用在屏蔽门系统上。
综上所述，本发明实施例提供了一种基于机器视觉技术的地铁站台门障碍物探测预警系统，通过采用以上以上技术方案，本发明实施例具有以下优点：
本发明的基于机器视觉技术的地铁站台门障碍物探测预警系统，使用机器视觉技术，具有三维属性，能够实时判断探测对象障碍物发生的时间，所属的种类和发生的位置等关键信息，能够综合评价障碍物是否具有危险性，并将具有危险性的障碍物或其行为解除在站台门和列车信号交接之前，其时效性、可靠性和安全性优于传统的探测预警系统。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详细描述的部分，可以参见其它实施例的相关描述。
上述实施例仅用以说明本发明的技术方案，而非对其限制；本领域的普通技术人员应当理解：其依然可以对上述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。
Claims (10)
- 一种基于机器视觉技术的地铁站台门障碍物探测预警系统，其特征在于，包括：摄像机，与摄像机通信连接的缓存设备，与缓存设备及摄像机通信连接的监控单元，以及与监控单元通信连接的报警装置；所述摄像机有两台，分别安装在地铁站台门两侧所设的机械结构上，且距离地面一定高度，并朝向地铁站台门的被探测区域。
- 根据权利要求1所述的系统，其特征在于，所述机械结构包括：固定在地面上的摄像机安装支架，和可调节的安装在摄像机安装支架上端的摄像机安装调节机构，所述摄像机安装调节机构上设有摄像机安装位。
- 根据权利要求2所述的系统，其特征在于，所述机械结构还包括安装在摄像机安装支架上端、用于保护摄像机的摄像机外防护罩和透明护板，所述摄像机外防护罩将整个摄像机容纳在其内部保护空间内，所述透明护板位于摄像机镜头前。
- 根据权利要求1所述的系统，其特征在于，所述地铁站台门包括位于两边的固定门单元和位于中间的滑动门单元，所述两台摄像机分别设置在两边的固定门单元的外侧，所述探测区域是指滑动门单元所形成的通道区域以及固定门单元与列车之间的间隙。
- 根据权利要求1-4中任一所述的系统，其特征在于，所述摄像机用于采集探测区域内障碍物的影像；所述缓存设备用于按时序储存采集的影像数据；所述监控单元用于分析处理采集的影像数据，识别障碍物类别，判断是否有危险，并在识别出危险时发出报警指令；所述报警装置用于在收到报警指令时进行报警。
- 根据权利要求5所述的系统，其特征在于，所述监控单元，还用于对来自缓存设备的影像数据进行分类整理，所述分类整理包括：按时序分类、按障碍物类别分类和按照危险性分类。
- 根据权利要求5所述的系统，其特征在于，所述监控单元还用于与地铁站台门控制单元通信连接并提供监控数据，以使地铁站台门控制单元结合得到的监控数据对地铁站台门实施控制。
- 根据权利要求5所述的系统，其特征在于，所述摄像机还与车站中央监控中心通信连接并提供监控数据。
- 根据权利要求8所述的系统，其特征在于，所述摄像机、所述缓存设备、所述监控单元和所述报警设备以及所述车站中央监控中心，共同构成障碍物探测识别系统；所述障碍物探测识别系统，用于进行列车进站前或出站前障碍物识别，如果存在障碍物，控制列车不能进站或出站；以及，用于进行列车停靠中障碍物识别，对站台门附近是否存在危险进行判断，根据判断结果执行相应操作。
- 根据权利要求5所述的系统，其特征在于，所述报警装置还用于发出列车进站预警、列车停靠预警和列车出站预警。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910105982.3A CN111516708A (zh) | 2019-02-01 | 2019-02-01 | 一种基于机器视觉技术的地铁站台门障碍物探测预警系统 |
| CN201910105982.3 | 2019-02-01 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020155294A1 true WO2020155294A1 (zh) | 2020-08-06 |
Family
ID=71841564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2019/077231 WO2020155294A1 (zh) | 2019-02-01 | 2019-03-06 | 一种基于机器视觉技术的地铁站台门障碍物探测预警系统 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111516708A (zh) |
| WO (1) | WO2020155294A1 (zh) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112406961B (zh) * | 2020-11-24 | 2023-01-24 | 南京康尼电子科技有限公司 | 站台门视频障碍物探测控制系统及方法 |
| CN112507998B (zh) * | 2021-02-08 | 2021-04-27 | 南京信息工程大学 | 基于机器视觉的屏蔽门行人等待提醒系统及其方法 |
| CN115320668B (zh) * | 2022-08-05 | 2023-08-18 | 卡斯柯信号有限公司 | 一种站台门状态检测系统及方法 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103091724A (zh) * | 2013-01-31 | 2013-05-08 | 华南理工大学 | 直线地铁站台屏蔽门与列车间异物自动检测方法及其装置 |
| CN104777521A (zh) * | 2015-03-24 | 2015-07-15 | 广州市地下铁道总公司 | 一种基于双目视觉的列车车门与站台屏蔽门之间异物检测系统及其检测方法 |
| CN105242614A (zh) * | 2015-11-17 | 2016-01-13 | 广州新科佳都科技有限公司 | 屏蔽门安全防护控制方法及系统 |
| CN105317310A (zh) * | 2014-07-14 | 2016-02-10 | 广州天赋人财光电科技有限公司 | 基于机器视觉的屏蔽门和列车门防夹超限监控系统及方法 |
| CN106864463A (zh) * | 2017-01-26 | 2017-06-20 | 广州地铁设计研究院有限公司 | 一种地铁站台门与车门之间顶置式异物检测系统 |
| CN108454635A (zh) * | 2018-03-31 | 2018-08-28 | 广州明森科技股份有限公司 | 一种地铁站台的屏蔽门与列车门之间的异物检测系统 |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4056813B2 (ja) * | 2002-07-11 | 2008-03-05 | 松下電器産業株式会社 | 障害物検知装置 |
| CN101898567B (zh) * | 2010-04-07 | 2012-03-14 | 西南交通大学 | 基于智能视频的铁路异物侵限监控系统 |
| KR101066562B1 (ko) * | 2011-02-15 | 2011-09-21 | 지에스네오텍(주) | 스테레오 영상을 이용한 스크린 도어 시스템 |
| CN104847211B (zh) * | 2015-03-24 | 2017-01-25 | 江苏科技大学 | 一种站台安全门与车门区间安全的辅助系统及其实现方法 |
| CN207241692U (zh) * | 2017-09-28 | 2018-04-17 | 天津天乐泰力科技有限公司 | 一种用于列车与站台屏蔽门之间间隙的视频检测系统 |
| CN107672625B (zh) * | 2017-10-10 | 2019-12-10 | 深圳市信海通科技有限公司 | 高速铁路周界入侵报警系统 |
| CN209505732U (zh) * | 2019-02-01 | 2019-10-18 | 方大智创科技有限公司 | 一种基于机器视觉技术的地铁站台门障碍物探测预警系统 |
-
2019
- 2019-02-01 CN CN201910105982.3A patent/CN111516708A/zh not_active Withdrawn
- 2019-03-06 WO PCT/CN2019/077231 patent/WO2020155294A1/zh active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103091724A (zh) * | 2013-01-31 | 2013-05-08 | 华南理工大学 | 直线地铁站台屏蔽门与列车间异物自动检测方法及其装置 |
| CN105317310A (zh) * | 2014-07-14 | 2016-02-10 | 广州天赋人财光电科技有限公司 | 基于机器视觉的屏蔽门和列车门防夹超限监控系统及方法 |
| CN104777521A (zh) * | 2015-03-24 | 2015-07-15 | 广州市地下铁道总公司 | 一种基于双目视觉的列车车门与站台屏蔽门之间异物检测系统及其检测方法 |
| CN105242614A (zh) * | 2015-11-17 | 2016-01-13 | 广州新科佳都科技有限公司 | 屏蔽门安全防护控制方法及系统 |
| CN106864463A (zh) * | 2017-01-26 | 2017-06-20 | 广州地铁设计研究院有限公司 | 一种地铁站台门与车门之间顶置式异物检测系统 |
| CN108454635A (zh) * | 2018-03-31 | 2018-08-28 | 广州明森科技股份有限公司 | 一种地铁站台的屏蔽门与列车门之间的异物检测系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111516708A (zh) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020155294A1 (zh) | 一种基于机器视觉技术的地铁站台门障碍物探测预警系统 | |
| WO2021159604A1 (zh) | 一种轨道列车的监控系统，监控方法及监控装置 | |
| JP4595010B2 (ja) | エスカレーター | |
| US20070291985A1 (en) | Intelligent railyard monitoring system | |
| CN101490283B (zh) | 具有涉及安全的区域的工业设备 | |
| KR102554662B1 (ko) | 이동식 무인탐지장치를 이용한 인공지능 기반 안전관리 시스템 | |
| KR20080044812A (ko) | 컴퓨터 영상 분석기술을 이용한 범죄 및 사고예방 자동경비 시스템 | |
| CN110053651A (zh) | 铁路列车检修区域内平交道口的安全防护系统及方法 | |
| CN103400430A (zh) | 基于视频光幕的地铁门安全监控系统 | |
| CN109658652A (zh) | 铁路车站站台立体化防护系统 | |
| CN202189450U (zh) | 民航机场安防监控指挥系统 | |
| CN105929772A (zh) | 掘进作业的红外成像人体识别安全预警装置 | |
| KR102008966B1 (ko) | 철도 승강장 추락사고 대응 시스템 | |
| CN209505732U (zh) | 一种基于机器视觉技术的地铁站台门障碍物探测预警系统 | |
| KR20040023786A (ko) | 열차운행 안전관리시스템 | |
| CN112406961B (zh) | 站台门视频障碍物探测控制系统及方法 | |
| KR102276635B1 (ko) | 고정 설치식 전동차 이상 검지 시스템 및 그 방법 | |
| CN211427457U (zh) | 一种重要处所智能安防系统 | |
| JP2001026266A (ja) | 軌道上障害物監視通報システム | |
| KR101266015B1 (ko) | 승강장 안전선 및 선로 감시 시스템 | |
| JP6149357B2 (ja) | 防護発報無線システム | |
| KR200459886Y1 (ko) | 교통사고 자동 기록장치 | |
| KR101012738B1 (ko) | 철도 역사의 테러 감시 시스템 및 방법 | |
| CN112037484A (zh) | 一种铁路货场防侵限安全智能预警系统 | |
| CN202189444U (zh) | 火车站智能化安防监控指挥系统 |
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: 19912270 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19912270 Country of ref document: EP Kind code of ref document: A1 |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19912270 Country of ref document: EP Kind code of ref document: A1 |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 21.01.2022) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19912270 Country of ref document: EP Kind code of ref document: A1 |