High-speed motor train unit rail foreign matter invasion risk monitoring method and related device

Abstract

The invention discloses a high-speed train rail foreign matter invasion risk monitoring method and a related device, and relates to the field of artificial intelligence. A first video image is acquired in real time through a camera; marking the first video image through a semantic segmentation model, and obtaining a second video image marked with a position area of the boundary of the rail and a position area of the railway overhead line system pillar; and marking the second video image through the target detection model to obtain a third video image marked with the rectangular frame of the biological object. The first position relation between the rectangular frame of the biological object and the position area of the boundary of the rail and the second position relation between the rectangular frame of the biological object and the position area of the railway overhead line system supporting column are determined. And determining the risk level of the biological object invading the limit orbit based on the information. According to the method provided by the embodiment of the invention, human intervention is not needed, and the risk of the biological object invading the limited orbit can be automatically monitored.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and related device for monitoring the risk of foreign matter intrusion on tracks of high-speed EMUs. Background Art

[0002] With the advancement of science and technology, high-speed EMUs have achieved great success and their coverage is getting wider and wider. The driving safety of high-speed EMUs is very important. Once a major traffic accident occurs, life and property will suffer huge losses. The main factor affecting the driving safety of high-speed EMUs is the intrusion of foreign objects on the track.

[0003] In the related technologies, there are two main methods to prevent foreign objects from intruding on the track: contact foreign object detection and non-contact foreign object detection. Contact foreign object detection is to detect whether foreign objects have intruded on the track by directly or indirectly contacting the sensor element with the foreign object, such as optical fiber cable detection and power grid detection. Non-contact foreign object detection is to detect whether foreign objects have intruded on the track by using the transmission characteristics of sound waves and electromagnetic waves, such as ultrasonic detection, radar detection, etc.

[0004] The above methods in the related art all require manual assistance and are cumbersome. Therefore, there is an urgent need for an automated high-speed EMU track foreign body intrusion risk monitoring method. Summary of the invention

[0005] In view of the above problems, the present application provides a method and related device for monitoring the risk of foreign objects intruding on the track of a high-speed EMU, so as to achieve the purpose of an automated method for monitoring the risk of foreign objects intruding on the track of a high-speed EMU. The specific scheme is as follows:

[0006] The first aspect of the present application provides a method for monitoring the risk of foreign matter intrusion on the track of a high-speed train, comprising:

[0007] Acquire a first video image captured in real time by a camera, wherein the first video image includes rails and railway contact network pillars;

[0008] Inputting the first video image into a pre-built semantic segmentation model, obtaining a second video image through the semantic segmentation model, wherein the second video image is marked with a location area of ​​a rail boundary and a location area of ​​a railway contact network support;

[0009] Inputting the second video image into a pre-built target detection model, obtaining a third video image through the target detection model, wherein the third video image is marked with a rectangular frame of the biological object, a location area of ​​the boundary of the rail, and a location area of ​​the railway contact network pillar;

[0010] Based on the plurality of frames of the third video images sorted from earliest to latest according to the acquisition time, determining the movement trajectory of the biological object, the first positional relationship between the rectangular frame of the biological object and the positional area of ​​the boundary of the rail, and the second positional relationship between the rectangular frame of the biological object and the positional area of ​​the railway contact network pillar;

[0011] Based on the movement trajectory of the biological object, the first position relationship, and the second position relationship, a risk level of the biological object intruding into a restricted track is determined.

[0012] In a possible implementation, the method further includes:

[0013] From the correspondence between the preset time when the high-speed train group arrives at the camera and the speed of the high-speed train group, find the target speed of the first high-speed train group corresponding to the target time later than the current time;

[0014] Determine that the product of the difference between the target time and the current time and the target speed is the real-time displacement between the first high-speed train group and the camera;

[0015] If there is a second high-speed train group in the first high-speed train group whose real-time displacement with respect to the camera is less than or equal to the preset displacement, the camera, the semantic segmentation model and the target detection model are controlled to be in working state.

[0016] In a possible implementation, the method further includes:

[0017] From the correspondence between the preset numbers of the high-speed train sets and the communication addresses of the high-speed train sets, searching for the first communication address corresponding to the number of the second high-speed train set;

[0018] Acquire a pre-stored second communication address of a monitoring device belonging to the dispatching command and control center;

[0019] The risk level is sent to the first communication address and the second communication address.

[0020] In a possible implementation, the step of determining the risk of the biological object invading a restricted track based on the movement trajectory of the biological object, the first position relationship, and the second position relationship includes:

[0021] If the second positional relationship indicates that the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object indicates that the biological object remains outside the position area of ​​the railway contact network pillar or the movement trajectory of the biological object indicates that the biological object is away from the position area of ​​the railway contact network pillar, the risk level of the biological object intruding the limited track is determined to be the first risk level;

[0022] If the second positional relationship represents that the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object represents that the biological object is close to the position area of ​​the railway contact network pillar, the risk level of the biological object invading the restricted track is determined to be the second risk level;

[0023] If the second positional relationship represents that the rectangular frame of the biological object is located in the first area, and the movement trajectory of the biological object represents that the biological object remains located in the first area or the movement trajectory of the biological object represents that the biological object is away from the location area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the third risk level; the first area is the area between the inner side of the location area of ​​the railway contact network pillar and the outer side of the location area of ​​the boundary of the railway;

[0024] If the second positional relationship represents that the rectangular frame of the biological object is located in the first area, and the movement trajectory of the biological object represents that the biological object is close to the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fourth risk level;

[0025] If the second positional relationship represents that the rectangular frame of the biological object is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object is far away from the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fourth risk level;

[0026] If the second positional relationship represents that the rectangular frame of the biological object is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object remains inside the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fifth risk level.

[0027] A second aspect of the present application provides a high-speed train track foreign object intrusion risk monitoring device, comprising:

[0028] A first acquisition module is used to acquire a first video image captured by a camera in real time, wherein the first video image includes rails and railway contact network pillars;

[0029] A second acquisition module is used to input the first video image into a pre-built semantic segmentation model, and acquire a second video image through the semantic segmentation model, wherein the second video image is marked with a location area of ​​a rail boundary and a location area of ​​a railway contact network support;

[0030] A third acquisition module is used to input the second video image into a pre-built target detection model, and acquire a third video image through the target detection model, wherein the third video image is marked with a rectangular frame of the biological object, a location area of ​​the boundary of the rail, and a location area of ​​the railway contact network pillar;

[0031] A first determination module is used to determine the movement trajectory of the biological object, the first position relationship between the rectangular frame of the biological object and the position area of ​​the boundary of the rail, and the second position relationship between the rectangular frame of the biological object and the position area of ​​the railway contact network pillar based on multiple frames of the third video image sorted from early to late according to the acquisition time;

[0032] The second determination module is used to determine the risk level of the biological object invading the restricted track based on the movement trajectory of the biological object, the first position relationship and the second position relationship.

[0033] In a possible implementation, the method further includes:

[0034] A first search module is used to search for a target speed of the first high-speed train group corresponding to a target time later than the current time from the correspondence between the preset time when the high-speed train group arrives at the camera and the speed of the high-speed train group;

[0035] A third determination module is used to determine that the product of the difference between the target time and the current time and the target speed is the real-time displacement of the first high-speed train group and the camera;

[0036] A control module is used to control the camera, the semantic segmentation model and the target detection model to be in a working state if there is a second high-speed train group in the first high-speed train group whose real-time displacement with the camera is less than or equal to the preset displacement.

[0037] In a possible implementation, the method further includes:

[0038] A second search module is used to search for the first communication address corresponding to the number of the second high-speed train group from the correspondence between the preset numbers of the high-speed train groups and the communication addresses of the high-speed train groups;

[0039] A fourth acquisition module, used to acquire a pre-stored second communication address of a monitoring device belonging to the dispatching command and control center;

[0040] A sending module is used to send the risk level to the first communication address and the second communication address.

[0041] The third aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on a roadside computing unit, the roadside computing unit implements the method for monitoring the risk of foreign object intrusion on the track of a high-speed EMU according to the first aspect or any implementation method of the first aspect.

[0042] A fourth aspect of the present application provides a roadside computing unit, comprising at least one processor and a memory connected to the processor, wherein:

[0043] The memory is used to store computer programs;

[0044] The processor is used to execute the computer program so that the roadside computing unit can implement the high-speed EMU track foreign object intrusion risk monitoring method of the above-mentioned first aspect or any implementation method of the first aspect.

[0045] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by a roadside computing unit, the roadside computing unit can implement the high-speed EMU track foreign object intrusion risk monitoring method according to the first aspect or any implementation of the first aspect.

[0046] By means of the above technical solution, the present application provides a method for monitoring the risk of foreign objects intruding on the track of a high-speed EMU, wherein a first video image is acquired in real time by a camera; the first video image is annotated by a semantic segmentation model to obtain a second video image annotated with the position area of ​​the rail boundary and the position area of ​​the railway contact network pillar; the second video image is annotated by a target detection model to obtain a third video image annotated with a rectangular frame of a biological object; then the third video image is annotated with a rectangular frame of a biological object, the position area of ​​the rail boundary and the position area of ​​the railway contact network pillar. Therefore, the moving trajectory of the biological object, the first position relationship between the rectangular frame of the biological object and the position area of ​​the rail boundary, and the second position relationship between the rectangular frame of the biological object and the position area of ​​the railway contact network pillar can be determined. Based on the moving trajectory, the first position relationship and the second position relationship of the biological object, the risk level of biological object intrusion on the track is determined. The method provided in the embodiment of the present application does not require human intervention and can automatically monitor the risk of biological objects intruding on the track. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0048] Figure 1A schematic diagram of a system architecture provided for this application;

[0049] Figure 2 A schematic diagram of an implementation method of the roadside assembly provided in an embodiment of the present application being arranged on a railway contact network pillar;

[0050] Figure 3 A schematic diagram of an optional hardware structure of a terminal provided in an embodiment of the present application;

[0051] Figure 4 A schematic diagram of the structure of a roadside calculation unit 100 provided in an embodiment of the present application;

[0052] Figure 5 A schematic diagram of a flow chart of a method for monitoring the risk of foreign matter intrusion on the track of a high-speed train provided in an embodiment of the present application;

[0053] Figure 6 A schematic diagram of comparison between a first video image and a second video image provided in an embodiment of the present application;

[0054] Figure 7 A schematic diagram of comparison between the second video image and the third video image provided in an embodiment of the present application;

[0055] Figure 8 A schematic diagram of the structure of a high-speed EMU track foreign object intrusion risk monitoring device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0057] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0058] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0059] See also Figure 1 , Figure 1 A schematic diagram of a system architecture is shown. The system may include a roadside computing unit 100, a camera 200, a terminal device 300 belonging to a high-speed train group, and a monitoring device 400 belonging to a dispatching command and control center.

[0060] For example, one or more cameras and a roadside computing unit 100 may constitute a roadside set; the roadside set may be set at a corresponding position on a railway contact network pillar. For example, the position of the camera installed on the railway contact network pillar may be determined based on actual conditions and is not limited here. For example, the position of the camera installed on the railway contact network pillar is the best position for the camera to capture the field of view.

[0061] Exemplarily, one or more roadside assemblies may be provided on a railway contact network pillar.

[0062] Exemplarily, the roadside assembly can be set on any device located on one side of the railway, and is not limited to being set on a railway contact network pillar.

[0063] In order for those skilled in the art to better understand the railway overhead line support and roadside assembly in the embodiment of the present application, the following is a detailed description. Figure 2 As shown, it is a schematic diagram of an implementation method of setting the roadside assembly provided in an embodiment of the present application on a railway contact network pillar.

[0064] Figure 2 The support used to erect high-voltage wires is called a railway contact network support 21.

[0065] Figure 2 The location of the roadside collection is set only as an example and is not limited in the embodiments of the present application.

[0066] Figure 2 The boundary of the railroad is still shown in the figure, such as the boundary 22 of the railroad. Figure 2 There are two tracks in it, so there are 4 rail boundaries.

[0067] It is understandable that the number of railroad boundaries in each frame of video image captured in real time by the camera 200 may be multiple.

[0068] like Figure 2 As shown, in the embodiment of the present application, the inside of the rail covered when the high-speed EMU is running is referred to as the inner side of the boundary of the rail, and the uncovered area is referred to as the outer side of the boundary of the rail.

[0069] In the embodiment of the present application, two side-by-side railway contact network pillars are referred to as a group of railway contact network pillars; the interior of a group of railway contact network pillars is referred to as the inner side of the railway contact network pillars, and the exterior of a group of railway contact network pillars is referred to as the outer side of the railway contact network pillars.

[0070] Exemplarily, the terminal device 300 belonging to the high-speed train group may be located in the driver's cab of the high-speed train group.

[0071] The terminal device 300 or the monitoring device 400 in the embodiment of the present application can be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiment of the present application does not impose any restrictions on this.

[0072] from Figure 2 It can be seen that multiple groups of railway contact network pillars can be arranged on both sides of the rails. For example, each group of railway contact network pillars can be set with a roadside set. For example, the roadside sets are arranged at intervals on the railway contact network pillars. For example, if multiple groups of railway contact network pillars are arranged in order from near to far, the first group of railway contact network pillars, the third group of railway contact network pillars, the fifth group of railway contact network pillars, ..., are all provided with roadside sets; the second group of railway contact network pillars, the fourth group of railway contact network pillars, ..., are not provided with roadside sets.

[0073] For example, it can be determined which group of railway overhead line pillars need to be set up as a roadside set based on the maximum acquisition area of ​​the camera. Figure 2 The “dash-dot line” shown represents the distance between two adjacent sets of railway contact network pillars.

[0074] For example, the maximum collection area of ​​a camera is the area within 450 meters from the camera. If the distance between two adjacent groups of railway contact network pillars is less than 450 meters and the distance between three adjacent groups of railway contact network pillars is greater than 450 meters, then each group of railway contact network pillars needs to set up a roadside collection.

[0075] In the embodiment of the present application, the roadside computing unit 100 can obtain the video image collected in real time by the camera 200. The roadside computing unit 100 stores a pre-built semantic segmentation model and a pre-built target detection model, and can implement the high-speed EMU track foreign object intrusion risk monitoring method provided in the embodiment of the present application, thereby sending the risk level of biological object intrusion to the terminal device 300 and the monitoring device 400. This enables the driver in the cab of the high-speed EMU or the staff of the dispatching command and control center to make decisions to reduce the collision between the high-speed EMU and the biological object.

[0076] In the embodiment of the present application, the terminal device 300 or the monitoring device 400 is referred to as a terminal, and the manifestation of the terminal is described below.

[0077] Figure 3 An optional hardware structure diagram of a terminal is shown.

[0078] refer to Figure 3 As shown, the terminal may include a radio frequency unit 110, a memory 120, an input unit 130, a display unit 140, a camera 150 (optional), an audio circuit 160 (optional), a speaker 161 (optional), a microphone 162 (optional), an earphone jack 163 (optional), a processor 170, an external interface 180, a power supply 190 and other components. Those skilled in the art will appreciate that Figure 3 This is merely an example of a terminal and does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0079] The input unit 130 can be used to receive input digital or character information, and generate key signal input related to the user settings and function control of the terminal. Specifically, the input unit 130 may include a touch screen 131 (optional) and / or other input devices 132. The touch screen 131 can collect the user's touch operations on or near it (such as the user's operation on or near the touch screen using any suitable object such as fingers, joints, stylus, etc.), and drive the corresponding connection device according to a pre-set program. The touch screen can detect the user's touch action on the touch screen, convert the touch action into a touch signal and send it to the processor 170, and can receive and execute the command sent by the processor 170; the touch signal at least includes the touch point coordinate information. The touch screen 131 can provide an input interface and an output interface between the terminal and the user. In addition, the touch screen can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic wave. In addition to the touch screen 131, the input unit 130 can also include other input devices. Specifically, other input devices 132 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.

[0080] Among them, the input device 132 can receive input data and the like.

[0081] The display unit 140 may be used to display information input by a user or provided to a user, various menus of the terminal, an interactive interface, file display, and / or playback of any multimedia file.

[0082] The memory 120 can be used to store instructions and data. The memory 120 can mainly include an instruction storage area and a data storage area. The data storage area can store various data, such as multimedia files, texts, etc.; the instruction storage area can store software units such as operating systems, applications, instructions required for at least one function, or their subsets and extensions. It can also include a non-volatile random access memory; provide the processor 170 with hardware, software and data resources including management of computing and processing equipment, and support control software and applications. It is also used for the storage of multimedia files, and the storage of running programs and applications.

[0083] The processor 170 is the control center of the terminal. It uses various interfaces and lines to connect various parts of the entire terminal. By running or executing instructions stored in the memory 120 and calling data stored in the memory 120, it executes various functions of the terminal and processes data, thereby controlling the terminal device as a whole. Optionally, the processor 170 may include one or more processing units; preferably, the processor 170 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application program, and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 170. In some embodiments, the processor and the memory may be implemented on a single chip, and in some embodiments, they may also be implemented separately on separate chips. The processor 170 may also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory 120, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.

[0084] The radio frequency unit 110 (optional) can be used for receiving and sending information or receiving and sending signals during a call, for example, after receiving the downlink information of the base station, it is sent to the processor 170 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, etc. In addition, the radio frequency unit 110 can also communicate with network devices and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (Wideband Code Division Multiple Access, WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (SMS), etc.

[0085] In this embodiment of the present application, the radio frequency unit 110 can send data to the roadside calculation unit 100 and receive the processing results sent by the roadside calculation unit 100.

[0086] It should be understood that the radio frequency unit 110 is optional and can be replaced by other communication interfaces, such as a network port.

[0087] The terminal also includes a power supply 190 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 170 through a power management system, so that the power management system can manage charging, discharging, power consumption and other functions.

[0088] The terminal also includes an external interface 180, which may be a standard Micro USB interface or a multi-pin connector, and may be used to connect the terminal to communicate with other devices, or to connect a charger to charge the terminal.

[0089] Although not shown, the terminal may also include a flashlight, a wireless fidelity (WiFi) module, a Bluetooth module, sensors with different functions, etc., which will not be described in detail here. Some or all of the methods described below may be applied in the following embodiments. Figure 3 in the terminal shown.

[0090] Next describe Figure 1 The product form of the middle road side calculation unit 100;

[0091] Figure 4 A schematic diagram of the structure of a roadside calculation unit 100 is provided, such as Figure 4 As shown, the roadside computing unit 100 includes a bus 401, a processor 402, a communication interface 403, and a memory 404. The processor 402, the memory 404, and the communication interface 403 communicate with each other via the bus 401.

[0092] The bus 401 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0093] The processor 402 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0094] The memory 404 may include a volatile memory, such as a random access memory (RAM). The memory 404 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0095] Among them, the memory 404 can be used to store software codes related to the high-speed EMU track foreign object intrusion risk monitoring method, the processor 402 can execute the steps of the high-speed EMU track foreign object intrusion risk monitoring method of the chip, and can also dispatch other units to realize corresponding functions.

[0096] It should be understood that the above-mentioned terminal and roadside computing unit 100 can be centralized or distributed devices, and the processors in the above-mentioned terminal and roadside computing unit 100 (such as processor 170 and processor 402) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (digital signal processing, DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with an instruction execution function, such as a CPU, DSP, etc., or a hardware system without an instruction execution function, such as an ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without an instruction execution function and hardware systems with an instruction execution function.

[0097] Reference Figure 5 , Figure 5 A flow chart of a method for monitoring the risk of foreign matter intrusion on a high-speed train track provided in an embodiment of the present application is shown in FIG. Figure 5 As shown, a method for monitoring the risk of foreign objects intruding on the track of a high-speed EMU provided in an embodiment of the present application may include steps S501 to S505, and these steps are described in detail below.

[0098] Step S501: Acquire a first video image captured by a camera in real time, wherein the first video image includes rails and railway contact network pillars.

[0099] Step S502: input the first video image into a pre-built semantic segmentation model, and obtain a second video image through the semantic segmentation model.

[0100] The second video image is marked with the location area of ​​the rail boundary and the location area of ​​the railway contact network pillar.

[0101] It is understandable that as time goes by, more and more first video images are captured by the camera, and then more and more second video images are processed by the semantic segmentation model.

[0102] The first video image and the second video image are described below with reference to examples.

[0103] like Figure 6 , which is a schematic diagram for comparing the first video image and the second video image provided in an embodiment of the present application.

[0104] Figure 6The upper side is the first video image. After the first video image is input into the semantic segmentation model, the semantic segmentation model can output Figure 6 Second video image on the lower side.

[0105] like Figure 6 As shown, the position area of ​​the boundary of the rail in the second video image is marked with a dotted frame. The position area of ​​the railway contact network support in the second video image is marked with a solid frame.

[0106] Exemplarily, the semantic segmentation model is trained by taking a sample image as input and taking the position area of ​​the rail boundary annotated in the sample image and the position area of ​​the railway contact network pillar annotated as training targets.

[0107] For example, the sample image may be labeled by using labelme software, or by other methods, which is not limited in the embodiments of the present application.

[0108] When labeling the sample image using labelme software, you can set the label.py file of Labelme and define the color of the location area of ​​the rail boundary as red and the color of the location area of ​​the railway contact network pillar as green. The labeled location area is a polygon. The positions of the vertices of the polygons of the location area of ​​the rail boundary (polygon) and the location area of ​​the railway contact network pillar (polygon) of the labeled sample image can be set to a json format file.

[0109] For example, the sample images annotated by the labelme software and the Railsem19 public dataset (including images of location areas marked with rail boundaries and location areas of railway contact network pillars) can be converted into a unified format and then used as sample images for training the semantic segmentation model.

[0110] Step S503: input the second video image into a pre-built target detection model, and obtain a third video image through the target detection model.

[0111] The third video image is marked with a rectangular frame of the biological object, the location area of ​​the boundary of the rail, and the location area of ​​the railway contact network pillar.

[0112] Exemplarily, biological objects include but are not limited to humans, animals and other organisms.

[0113] The second video image and the third video image are described below with reference to examples.

[0114] like Figure 7 , which is a schematic diagram for comparing the second video image and the third video image provided in an embodiment of the present application.

[0115] Figure 7 The upper side is the second video image. After the second video image is input into the target detection model, the target detection model can output Figure 7 The third video image on the lower side.

[0116] like Figure 7 As shown, the rectangular frame of the biological object in the third video image is marked with a solid line frame.

[0117] Exemplarily, the target detection model is trained by taking a sample image as input and a rectangular box of a biological object annotated in the sample image as a training target.

[0118] For example, the sample image may be labeled by using labelme software, or by other methods, which is not limited in the embodiments of the present application.

[0119] For example, in the process of labeling the sample image by using labelme software, the label.py file of Labelme can be set. The labeled position area is a polygon. The positions of the vertices of the rectangular frame (polygon) of the biological object of the labeled sample image can be set to a file in json format.

[0120] Exemplarily, sample images annotated by labelme software and KITTI public datasets (including images annotated with rectangular frames of biological objects) can be converted into a unified format and then used as sample images for training the target detection model.

[0121] Step S504: Based on multiple frames of the third video image sorted from early to late according to the acquisition time, determine the movement trajectory of the biological object, the first position relationship between the rectangular frame of the biological object and the position area of ​​the boundary of the rail, and the second position relationship between the rectangular frame of the biological object and the position area of ​​the railway contact network pillar.

[0122] Exemplarily, the acquisition time of the third video image is the acquisition time of the first video image corresponding to the third video image acquired by the camera.

[0123] Exemplarily, the first position relationship includes: the rectangular frame of the biological object is located inside the position area of ​​the boundary of the railroad track (ie Figure 2 The biological object is located inside the boundary of the railroad track) or the rectangular frame of the biological object is outside the location area of ​​the railroad track boundary (i.e. Figure 2 outside the boundary of the railroad tracks in the ).

[0124] Exemplarily, the second positional relationship includes: the rectangular frame of the biological object is located inside the position area of ​​the railway contact network pillar (i.e. Figure 2The inner side of the railway contact network pillar in the figure) or the rectangular frame of the biological object is located outside the location area of ​​the railway contact network pillar (i.e. Figure 2 The outside of the railway overhead contact network pillar).

[0125] Step S505: Determine the risk level of the biological object intruding into a restricted track based on the movement trajectory of the biological object, the first position relationship, and the second position relationship.

[0126] The embodiment of the present application provides a method for monitoring the risk of foreign objects intruding on the track of a high-speed EMU, which obtains a first video image in real time through a camera; annotates the first video image through a semantic segmentation model to obtain a second video image with a position area marked with the boundary of the rail and the position area of ​​the railway contact network pillar; annotates the second video image through a target detection model to obtain a third video image with a rectangular frame marked with a biological object; then the third video image is marked with a rectangular frame of the biological object, the position area of ​​the boundary of the rail, and the position area of ​​the railway contact network pillar. Therefore, the moving trajectory of the biological object, the first position relationship between the rectangular frame of the biological object and the position area of ​​the boundary of the rail, and the second position relationship between the rectangular frame of the biological object and the position area of ​​the railway contact network pillar can be determined. Based on the moving trajectory of the biological object, the first position relationship and the second position relationship, the risk level of biological object intrusion on the track is determined. The method provided in the embodiment of the present application does not require human intervention and can automatically realize the monitoring of the risk of biological object intrusion on the track.

[0127] It is understandable that the camera and the roadside computing unit can be in working state all the time. If no high-speed train passes by the camera, or the high-speed train is far away from the camera, such as 300 km, then the operation of the camera and the roadside computing unit will cause a waste of resources. Based on this, the embodiment of the present application also provides the following method, which includes the following steps A1 to A3.

[0128] Step A1: From the preset correspondence between the time when the high-speed train arrives at the camera and the speed of the high-speed train, find the target speed of the first high-speed train corresponding to the target time later than the current time.

[0129] Assuming that the current time is 14:49 on November 13, 2024, the target time later than the current time may be any time later than 14:49 on November 13, 2024.

[0130] It is understandable that some high-speed trains may be delayed or their running time may be changed, so the equipment belonging to the dispatching command and control center can update the time when the high-speed train arrives at the camera stored in the roadside computing unit in real time. For example, the speed of the high-speed train can be calculated at the maximum speed.

[0131] Step A2: Determine the product of the difference between the target time and the current time and the target speed as the real-time displacement of the first high-speed train set and the camera.

[0132] Step A3: If there is a second high-speed train group in the first high-speed train group whose real-time displacement with the camera is less than or equal to the preset displacement, control the camera, the semantic segmentation model and the target detection model to be in working state.

[0133] Exemplarily, the preset displacement may be determined based on actual conditions, which is not limited in the present application. For example, the preset displacement may be 1,100 meters.

[0134] In an optional implementation, after determining the risk level, the roadside computing unit needs to inform the monitoring equipment of the dispatching command and control center and the terminal equipment of the driver's cab of the high-speed train. The specific method includes the following steps B1 to B3.

[0135] Step B1: searching for the first communication address corresponding to the number of the second high-speed train set from the correspondence between the preset numbers of the high-speed train sets and the communication addresses of the high-speed train sets.

[0136] Step B2: Acquire the pre-stored second communication address of the monitoring device belonging to the dispatching command and control center.

[0137] Step B3: Send the risk level to the first communication address and the second communication address.

[0138] It is understandable that after seeing the risk level, the staff of the dispatching command and control center or the driver in the cab of the high-speed EMU can respond in time to reduce the possibility of the high-speed EMU colliding with biological objects and derailing, thereby realizing real-time high-speed EMU operation line abnormality monitoring and early warning.

[0139] It is understandable that there are multiple situations in step S505, and the embodiments of the present application provide but are not limited to the following 6 situations.

[0140] Case 1: If the second positional relationship represents that the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object represents that the biological object remains outside the position area of ​​the railway contact network pillar or the movement trajectory of the biological object represents that the biological object is away from the position area of ​​the railway contact network pillar, the risk level of the biological object invading the limited track is determined to be the first risk level.

[0141] Combination Figure 2, if the rectangular frame of the biological object is located outside the location area of ​​the railway contact network pillar (such as Figure 6 The biological object is located outside the rail boundary (the left side of the railway contact network pillar 61 and the right side of the railway contact network pillar 62). Since the biological object is located outside the location area of ​​the railway contact network pillar, it means that even if a high-speed train passes by, it will not affect the biological object, and the biological object will not affect the high-speed train.

[0142] If the movement trajectory of the biological object indicates that the biological object is moving further and further away from the location area of ​​the railway contact network pillar, and the biological object is outside the location area of ​​the railway contact network pillar, it means that the biological object is moving further and further away from the railway contact network pillar and the rails, and the biological object and the high-speed EMU may not affect each other.

[0143] If the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object indicates that the biological object remains outside the position area of ​​the railway contact network pillar, it means that the biological object has no intention of approaching the railway contact network pillar and the rails, and the biological object and the high-speed train may not affect each other.

[0144] Case 2: If the rectangular frame represented by the second positional relationship as the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object represents the position area of ​​the biological object close to the railway contact network pillar, the risk level of the biological object invading the restricted track is determined to be the second risk level.

[0145] If the movement trajectory of the biological object indicates that the biological object is getting closer and closer to the location area of ​​the railway contact network pillar, and the biological object is outside the location area of ​​the railway contact network pillar, it means that the biological object is approaching the railway contact network pillar and the rail, that is, the biological object may enter the inner side of the location area of ​​the railway contact network pillar, and even the inner side of the location area of ​​the boundary of the rail. Therefore, the biological object may affect the operation of the high-speed EMU.

[0146] However, since the biological object has not yet entered the inner side of the location area of ​​the railway contact network pillar, it is possible to intervene in advance, for example, by informing the staff of the dispatching command and control center so that the staff can arrive at the scene in time and persuade the biological object to leave.

[0147] Exemplarily, a loudspeaker may be provided on a railway contact network pillar, and a corresponding voice may be played through the loudspeaker to persuade the biological object to leave.

[0148] Case three: If the second positional relationship represents that the rectangular frame of the biological object is located in the first area, and the movement trajectory of the biological object represents that the biological object remains located in the first area or the movement trajectory of the biological object represents that the biological object is away from the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the third risk level.

[0149] The first area is an area between the inner side of the location area of ​​the railway overhead wire support and the outer side of the location area of ​​the boundary of the railway.

[0150] If the biological object is located in the first area, and the movement trajectory of the biological object indicates that the biological object remains in the first area, it means that the biological object may not intend to enter the railroad. However, the biological object is close to the boundary of the railroad, so there is a certain risk.

[0151] If the biological object is located in the first area, and the movement trajectory of the biological object represents the location area of ​​the biological object far away from the boundary of the railway, it means that the biological object has the intention to leave the boundary of the railway track but has no intention to enter the railway track. However, the biological object is close to the boundary of the railway track, so there is also a certain risk.

[0152] Case 4: If the second positional relationship represents that the rectangular frame of the biological object is located in the first area, and the movement trajectory of the biological object represents the position area of ​​the biological object close to the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fourth risk level.

[0153] If the biological object is located in the first area, and the movement trajectory of the biological object represents a location area where the biological object is close to the boundary of the railway, it means that the biological object intends to enter the boundary of the railway track, and there is a relatively large risk.

[0154] Case 5: If the rectangular frame of the biological object represented by the second positional relationship is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object is away from the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fourth risk level.

[0155] If the biological object is already inside the location area of ​​the rail boundary, but has the intention to leave the location area of ​​the rail boundary, the risk level is still high because the biological object is located inside the location area of ​​the rail boundary.

[0156] Case six: If the second positional relationship represents that the rectangular frame of the biological object is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object remains inside the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fifth risk level.

[0157] If the biological object is already inside the location area of ​​the border of the railroad track, and the biological object has the intention to always stay inside the location area of ​​the border of the railroad track, the risk level is higher.

[0158] The semantic segmentation model and target detection model mentioned in the embodiments of the present application are explained below.

[0159] Exemplarily, the semantic segmentation model may be a deeplabv3+ algorithm model. The deeplabv3+ algorithm model is a lightweight multi-scale semantic segmentation algorithm, including an encoder module and a decoder module. The segmentation accuracy of the location area of ​​the rail boundary and the location area of ​​the railway contact network pillar is improved by fusing low-level spatial information with high-level semantic information.

[0160] After the first video image is input into the backbone (i.e., DCNN Atrous Conv), two outputs can be obtained: one output is the feature maps output by the last layer of convolution, and the other is the feature maps of the middle low-level features; the first part of the backbone output is sent to the ASPP module, and the second part is sent to the Decoder module. The Decoder module receives the low-level feature maps from the middle layer of the backbone and the output from the ASPP module as input, and fuses the two parts to improve the segmentation accuracy of the model.

[0161] Exemplarily, the target detection model may be a YOLOv5 algorithm model. The YOLOv5 algorithm model includes an input module, a Backbone module, a Neck network, and a Head module.

[0162] The YOLOv5 algorithm model slices the second video image of size 640×640×3 through the Backbone module to obtain a feature map of size 320×320×12, and then obtains a feature map of size 320×320×12 through 32 convolution kernels. Then, through the Neck network of the FPN+PAN structure, a feature map with multi-scale information is generated to improve the accuracy of target detection. The third video image is obtained through the Head module.

[0163] Exemplarily, the algorithm for determining the movement trajectory of the biological object in step S504 may be a DeepSort algorithm model. Exemplarily, multiple frames of the third video image may be sequentially input into the DeepSort algorithm model according to the acquisition time from early to late, and the movement trajectory of each biological object may be output through the DeepSort algorithm model.

[0164] The DeepSort algorithm module realizes continuous tracking of one or more biological objects by fusing the results output by the target detection model, combining Hungarian algorithm and Kalman filtering and other technologies. The DeepSort algorithm module cascade matches the rectangular frame of the biological features of the current video image with the rectangular frame of the confirmed movement trajectory in the previous frame of the video image, and uses the Hungarian algorithm to achieve the optimal linear allocation. The matched movement trajectory updates its state information according to the rectangular frame result of the current frame of the video image. The unmatched rectangular frame is IOU matched with the rectangular frame of the unconfirmed movement trajectory of the previous frame of the video image and the rectangular frame of the unmatched movement trajectory of the cascade matching. The unmatched rectangular frame creates a corresponding new movement trajectory and initializes the corresponding state and unconfirmed state; the unmatched movement trajectory updates its internal state according to its age; the matched movement trajectory updates its state according to the rectangular frame. The unmatched rectangular frame will be IOU matched with the rectangular frame of the unconfirmed movement trajectory in the previous frame of the video image and the rectangular frame of the unmatched movement trajectory of the cascade matching, and the Hungarian algorithm is used for linear allocation. The unmatched rectangular frame creates a corresponding new movement trajectory and initializes the corresponding state and unconfirmed state; the unmatched movement trajectory updates the internal state; the matched movement trajectory updates its state according to the detection frame. Finally, the motion parameters such as the speed, acceleration and trajectory of the biological object are obtained.

[0165] Exemplarily, the movement trajectory of the biological object includes motion parameters such as speed, acceleration and trajectory.

[0166] In order to help those skilled in the art better understand the embodiments of the present application, the applicant collected sample images at key locations along the railway line outside Yanshan University in Qinhuangdao and near the railway station, and annotated the sample images using labelme software, and combined open source data sets with relevant images crawled from the Internet to obtain annotated sample images. The hardware configuration parameters of the roadside computing unit developed using the Python3.8+PyTorch framework are set as shown in Table 1 below.

[0167] Table 1

[0168]

[0169] After the semantic segmentation model and the target detection model are trained using sample images, the semantic segmentation model and the target detection model are deployed in the roadside computing unit. The high-speed EMU track foreign object intrusion risk monitoring method provided in the embodiment of the present application is executed by the roadside computing unit, and the risk level of biological object intrusion into the track can be accurately determined.

[0170] The above introduces a method for monitoring the risk of foreign objects intruding on the track of a high-speed EMU provided in an embodiment of the present application. The following will introduce a device for executing the above-mentioned method for monitoring the risk of foreign objects intruding on the track of a high-speed EMU.

[0171] See also Figure 8 , Figure 8 This is a schematic diagram of the structure of a high-speed train track foreign body intrusion risk monitoring device provided in an embodiment of the present application. Figure 8 As shown, the high-speed EMU track foreign body intrusion risk monitoring device includes:

[0172] A first acquisition module 81 is used to acquire a first video image captured by a camera in real time, wherein the first video image includes rails and railway contact network pillars;

[0173] A second acquisition module 82 is used to input the first video image into a pre-built semantic segmentation model, and acquire a second video image through the semantic segmentation model, wherein the second video image is marked with a location area of ​​a rail boundary and a location area of ​​a railway contact network support;

[0174] A third acquisition module 83 is used to input the second video image into a pre-built target detection model, and acquire a third video image through the target detection model, wherein the third video image is marked with a rectangular frame of the biological object, a location area of ​​the boundary of the rail, and a location area of ​​the railway contact network pillar;

[0175] A first determination module 84 is used to determine the movement trajectory of the biological object, the first position relationship between the rectangular frame of the biological object and the position area of ​​the boundary of the rail, and the second position relationship between the rectangular frame of the biological object and the position area of ​​the railway contact network pillar based on the multiple frames of the third video image sorted from early to late according to the acquisition time;

[0176] The second determination module 85 is used to determine the risk level of the biological object invading the restricted track based on the movement trajectory of the biological object, the first position relationship and the second position relationship.

[0177] In an optional implementation, the method further includes:

[0178] A first search module is used to search for a target speed of the first high-speed train group corresponding to a target time later than the current time from the correspondence between the preset time when the high-speed train group arrives at the camera and the speed of the high-speed train group;

[0179] A third determination module is used to determine that the product of the difference between the target time and the current time and the target speed is the real-time displacement of the first high-speed train group and the camera;

[0180] A control module is used to control the camera, the semantic segmentation model and the target detection model to be in a working state if there is a second high-speed train group in the first high-speed train group whose real-time displacement with the camera is less than or equal to the preset displacement.

[0181] In an optional implementation, the method further includes:

[0182] A second search module is used to search for the first communication address corresponding to the number of the second high-speed train group from the correspondence between the preset numbers of the high-speed train groups and the communication addresses of the high-speed train groups;

[0183] A fourth acquisition module, used to acquire a pre-stored second communication address of a monitoring device belonging to the dispatching command and control center;

[0184] A sending module is used to send the risk level to the first communication address and the second communication address.

[0185] In an optional implementation, the second determining module includes:

[0186] A first determining unit is used to determine that the risk level of the biological object intruding the limited track is a first risk level if the second position relationship represents that the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object represents that the biological object remains outside the position area of ​​the railway contact network pillar or the movement trajectory of the biological object represents that the biological object is away from the position area of ​​the railway contact network pillar;

[0187] A second determining unit is used to determine that the risk level of the biological object intruding the restricted track is a second risk level if the rectangular frame represented by the second positional relationship as the biological object is located outside the position area of ​​the railway contact network pillar and the movement trajectory of the biological object represents the position area where the biological object is close to the railway contact network pillar;

[0188] A third determining unit is configured to determine that the risk level of the biological object invading the restricted track is a third risk level if the rectangular frame of the biological object represented by the second positional relationship is located in a first area, and the movement trajectory of the biological object represents that the biological object remains located in the first area or the movement trajectory of the biological object represents that the biological object is away from the location area of ​​the boundary of the railway; the first area is the area between the inner side of the location area of ​​the railway contact network pillar and the outer side of the location area of ​​the boundary of the railway;

[0189] a fourth determining unit, configured to determine that the risk level of the biological object intruding into the restricted track is a fourth risk level if the rectangular frame represented by the second positional relationship as the biological object is located in the first area and the movement trajectory of the biological object represents a location area of ​​the biological object close to the boundary of the railway;

[0190] a fifth determining unit, configured to determine that the risk level of the biological object intruding into the restricted track is a fourth risk level if the rectangular frame represented by the second positional relationship as the biological object is located inside the position area of ​​the boundary of the railway and the movement trajectory of the biological object represents that the biological object is away from the position area of ​​the boundary of the railway;

[0191] The sixth determination unit is used to determine that the risk level of the biological object intruding into the restricted track is the fifth risk level if the rectangular frame represented by the second position relationship as the biological object is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object remains inside the position area of ​​the boundary of the railway.

[0192] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on a roadside computing unit, the roadside computing unit implements any of the high-speed EMU track foreign object intrusion risk monitoring methods provided in the embodiments of the present application.

[0193] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by a roadside computing unit, the roadside computing unit can implement any of the high-speed EMU track foreign object intrusion risk monitoring methods provided in the embodiments of the present application.

[0194] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0195] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0196] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0197] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A method for monitoring the risk of foreign matter intrusion on high-speed train tracks, characterized in that: include: Acquire a first video image captured in real time by a camera, wherein the first video image includes rails and railway contact network pillars; Inputting the first video image into a pre-built semantic segmentation model, obtaining a second video image through the semantic segmentation model, wherein the second video image is marked with a location area of ​​a rail boundary and a location area of ​​a railway contact network support; Inputting the second video image into a pre-built target detection model, obtaining a third video image through the target detection model, wherein the third video image is marked with a rectangular frame of the biological object, a location area of ​​the boundary of the rail, and a location area of ​​the railway contact network pillar; Based on the plurality of frames of the third video images sorted from earliest to latest according to the acquisition time, determining the movement trajectory of the biological object, the first positional relationship between the rectangular frame of the biological object and the positional area of ​​the boundary of the rail, and the second positional relationship between the rectangular frame of the biological object and the positional area of ​​the railway contact network pillar; Based on the movement trajectory of the biological object, the first position relationship, and the second position relationship, a risk level of the biological object intruding into a restricted track is determined.

2. According to the method for monitoring the risk of foreign matter intrusion on the track of a high-speed train set in claim 1, it is characterized in that: Also includes: From the correspondence between the preset time when the high-speed train group arrives at the camera and the speed of the high-speed train group, find the target speed of the first high-speed train group corresponding to the target time later than the current time; Determine that the product of the difference between the target time and the current time and the target speed is the real-time displacement between the first high-speed train group and the camera; If there is a second high-speed train group in the first high-speed train group whose real-time displacement with respect to the camera is less than or equal to the preset displacement, the camera, the semantic segmentation model and the target detection model are controlled to be in working state.

3. According to claim 2, the method for monitoring the risk of foreign matter intrusion on the track of a high-speed train is characterized in that: Also includes: From the correspondence between the preset numbers of the high-speed train sets and the communication addresses of the high-speed train sets, searching for the first communication address corresponding to the number of the second high-speed train set; Acquire a pre-stored second communication address of a monitoring device belonging to the dispatching command and control center; The risk level is sent to the first communication address and the second communication address.

4. According to claim 1, the method for monitoring the risk of foreign matter intrusion on the track of a high-speed train is characterized in that: The step of determining the risk of the biological object invading a restricted track based on the moving track of the biological object, the first position relationship and the second position relationship comprises: If the second positional relationship indicates that the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object indicates that the biological object remains outside the position area of ​​the railway contact network pillar or the movement trajectory of the biological object indicates that the biological object is away from the position area of ​​the railway contact network pillar, the risk level of the biological object intruding the limited track is determined to be the first risk level; If the second positional relationship represents that the rectangular frame of the biological object is located outside the position area of ​​the railway contact network pillar, and the movement trajectory of the biological object represents that the biological object is close to the position area of ​​the railway contact network pillar, the risk level of the biological object invading the restricted track is determined to be the second risk level; If the second positional relationship represents that the rectangular frame of the biological object is located in the first area, and the movement trajectory of the biological object represents that the biological object remains located in the first area or the movement trajectory of the biological object represents that the biological object is away from the location area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the third risk level; the first area is the area between the inner side of the location area of ​​the railway contact network pillar and the outer side of the location area of ​​the boundary of the railway; If the second positional relationship represents that the rectangular frame of the biological object is located in the first area, and the movement trajectory of the biological object represents that the biological object is close to the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fourth risk level; If the second positional relationship represents that the rectangular frame of the biological object is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object is far away from the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fourth risk level; If the second positional relationship represents that the rectangular frame of the biological object is located inside the position area of ​​the boundary of the railway, and the movement trajectory of the biological object represents that the biological object remains inside the position area of ​​the boundary of the railway, the risk level of the biological object invading the restricted track is determined to be the fifth risk level.

5. A high-speed train track foreign body intrusion risk monitoring device, characterized in that: include: A first acquisition module is used to acquire a first video image captured by a camera in real time, wherein the first video image includes rails and railway contact network pillars; A second acquisition module is used to input the first video image into a pre-built semantic segmentation model, and acquire a second video image through the semantic segmentation model, wherein the second video image is marked with a location area of ​​a rail boundary and a location area of ​​a railway contact network support; A third acquisition module is used to input the second video image into a pre-built target detection model, and acquire a third video image through the target detection model, wherein the third video image is marked with a rectangular frame of the biological object, a location area of ​​the boundary of the rail, and a location area of ​​the railway contact network pillar; A first determination module is used to determine the movement trajectory of the biological object, the first position relationship between the rectangular frame of the biological object and the position area of ​​the boundary of the rail, and the second position relationship between the rectangular frame of the biological object and the position area of ​​the railway contact network pillar based on multiple frames of the third video image sorted from early to late according to the acquisition time; The second determination module is used to determine the risk level of the biological object invading the restricted track based on the movement trajectory of the biological object, the first position relationship and the second position relationship.

6. According to claim 5, the high-speed train track foreign body intrusion risk monitoring device is characterized in that: Also includes: A first search module is used to search for a target speed of the first high-speed train group corresponding to a target time later than the current time from the correspondence between the preset time when the high-speed train group arrives at the camera and the speed of the high-speed train group; A third determination module is used to determine that the product of the difference between the target time and the current time and the target speed is the real-time displacement of the first high-speed train group and the camera; A control module is used to control the camera, the semantic segmentation model and the target detection model to be in a working state if there is a second high-speed train group in the first high-speed train group whose real-time displacement with the camera is less than or equal to the preset displacement.

7. The high-speed train track foreign body intrusion risk monitoring device according to claim 6 is characterized in that: Also includes: A second search module is used to search for the first communication address corresponding to the number of the second high-speed train group from the correspondence between the preset numbers of the high-speed train groups and the communication addresses of the high-speed train groups; A fourth acquisition module, used to acquire a pre-stored second communication address of a monitoring device belonging to the dispatching command and control center; A sending module is used to send the risk level to the first communication address and the second communication address.

8. A computer program product, characterized in that It includes computer-readable instructions. When the computer-readable instructions are executed on a roadside computing unit, the roadside computing unit implements the high-speed EMU track foreign object intrusion risk monitoring method as described in any one of claims 1 to 4.

9. A roadside calculation unit, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the roadside computing unit can implement the high-speed EMU track foreign object intrusion risk monitoring method as described in any one of claims 1 to 4.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs. When the one or more computer programs are executed by a roadside computing unit, the roadside computing unit can implement the high-speed EMU track foreign object intrusion risk monitoring method as described in any one of claims 1 to 4.

Images