Urban rail transit straight line section line foreign matter invasion identification method

Abstract

The invention discloses an urban rail transit straight line section line foreign matter invasion identification method, and the method comprises the steps: arranging an image detection device and a radar detection device at the end part of an urban rail transit locomotive, and carrying out the data collection; then a train-mounted computing device is used for recognizing and detecting a real-time view image in the collected data, and a view image feature detection result is obtained; identifying and monitoring the real-time radar data to obtain a current radar detection foreign matter position; according to the data synchronization frequency, performing target position fusion on the current radar detection object position after coordinate conversion and the view image feature detection result to obtain a vehicle-mounted foreign matter invasion detection result; the vehicle-mounted invasion detection result is uploaded to the foreign matter invasion detection center system through the train-ground wireless communication system, the train position information K and the radar detection object distance delta x are summed in real time, foreign matter position information K + delta x is calculated, and in combination with the foreign matter invasion detection result, mapping alarm is conducted on a route diagram of the foreign matter invasion detection center system.

Description

Technical Field

[0001] The present invention belongs to the technical field of rail transit safety. More specifically, it relates to a method for identifying foreign object intrusion in the straight section of urban rail transit lines. Background Art

[0002] Foreign objects specifically refer to: any object (such as tools, materials, equipment parts, foreign items, etc.) or non-rail vehicles / machineries that enter the traffic limit range of the rail transit line, exceed the safe operation space, and may pose a risk of collision, derailment or other accidents. In recent years, urban rail transit accidents caused by foreign object intrusion have been common.

[0003] The elevated section of the urban rail transit line is vulnerable to foreign items being blown into the track by the wind (such as packaging bags, billboards, tree branches, etc.), off-road construction or vehicle falling objects invading the track safety protection area (such as unenclosed management of construction sites); inside the tunnel and vehicle depots, it is vulnerable to construction tools or materials left uncleaned or foreign objects left due to staff operation errors, which pose a threat to train operation safety. And the train runs at a high speed in the straight section. If there is a lack of an efficient and accurate foreign object intrusion detection and identification method, it is difficult to ensure the train operation safety. Currently, most of the existing urban rail transit lines detect obstacle intrusion by installing millimeter-wave radars, lidars, visual sensors, acoustic sensors, etc. on the ground. Although it can reduce the occurrence of missed detections of obstacles, its effective detection distance is short and there are blind spots, and there are still problems with insufficient accuracy and efficiency in obstacle recognition and positioning. If considering installing foreign object intrusion detection devices on the ground for full line coverage, the construction cost will also increase significantly. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art, provide a method for identifying foreign object intrusion in the straight section of urban rail transit lines, solve the problem that the existing urban rail transit foreign object intrusion detection is not comprehensive and accurate enough, and at the same time improve the economy of the system device.

[0005] To achieve the above-mentioned invention purpose, a method for identifying foreign object intrusion in urban rail transit lines of the present invention is characterized by including the following steps:

[0006] (1) Set up a foreign object intrusion detection center system for the straight section at the urban rail transit line control center, and set up a dedicated interface for train position information between the foreign object intrusion detection center system and the train control system to obtain the train position information K in real time;

[0007] (2) Set the image detection device and the radar detection device at the end of the locomotive to capture the real-time visual image in the train traveling direction and detect the real-time radar data in the train traveling direction in the straight section;

[0008] (3) Use the vision image foreign object detection neural network in the train on-vehicle computing device to identify and detect the real-time vision image, and obtain the vision image feature detection result;

[0009] (4) According to the Kalman filtering algorithm, use the train on-vehicle computing device to identify and monitor the real-time radar data, obtain the distance Δx between the currently detected radar object and the locomotive end, and then use the vehicle-ground wireless communication system to upload the radar detection object distance Δx to the foreign object intrusion detection center system in real time;

[0010] (5) According to the data synchronization frequency, use the train on-vehicle computing device to fuse the current radar detection object position and the vision image feature detection result for target position, and obtain the foreign object intrusion detection result;

[0011] (6) The foreign object intrusion detection center system creates a center system line map according to the mileage characteristics of the rail transit line, then maps the train's traveling position into the line map, establishes a real-time train position display, and sums the train position information K and the radar detection object distance Δx in real time to calculate the foreign object position information, and then performs a mapping alarm on the line map of the foreign object intrusion detection center system.

[0012] The invention purpose of the present invention is achieved as follows:

[0013] A method for identifying foreign object intrusion on a straight section line of urban rail transit of the present invention first establishes a foreign object intrusion detection system for the straight section line. This system consists of a foreign object intrusion detection center system and an on-vehicle foreign object intrusion detection subsystem. The on-vehicle foreign object intrusion detection subsystem includes: an image detection device, a radar detection device, and a train on-vehicle computing device. An image detection device and a radar detection device are set at the locomotive end of the urban rail transit for data collection; then use the train on-vehicle computing device to identify and detect the real-time vision image in the collected data to obtain the vision image feature detection result; identify and monitor the real-time radar data to obtain the current radar detection foreign object position; according to the data synchronization frequency, fuse the current radar detection object position after coordinate conversion and the vision image feature detection result for target position to obtain the on-vehicle foreign object intrusion detection result; upload the on-vehicle intrusion detection result to the foreign object intrusion detection center system through the existing vehicle-ground wireless communication system of the urban rail transit. The foreign object intrusion detection center system sums the train position information K and the radar detection object distance Δx in real time to calculate the foreign object position information K + Δx, and combines the foreign object intrusion detection result to perform a mapping alarm on the line map of the foreign object intrusion detection center system.

[0014] At the same time, a method for identifying foreign object intrusion on a straight section line of urban rail transit of the present invention also has the following beneficial effects:

[0015] (1). The present invention constructs an urban rail transit foreign object intrusion detection system by combining on-vehicle and ground center methods. The main application scenario of this system is the straight-line section of urban rail transit. Through the active detection method of the train, the installation quantity of ground foreign object detection devices is effectively reduced, thereby saving the equipment investment along the line. An economical and efficient detection method is proposed for the foreign object detection of the straight-line section of urban rail transit.

[0016] (2). The present invention uses a vision image foreign object detection neural network to identify and detect real-time vision images, processes radar data with the Kalman filtering algorithm, and based on the data synchronization frequency, fuses the position of the radar-detected object after coordinate conversion with the detection result of the vision image features, improving the accuracy and precision of foreign object detection;

[0017] (3). The present invention utilizes the existing vehicle-ground wireless communication system and train control system of urban rail transit. By simulating and constructing a track line map, warning information and train position information can be mapped on the line map, realizing the visual management of foreign object intrusion, facilitating an intuitive understanding of the situation of foreign object intrusion, quickly making decisions and taking corresponding countermeasures, and effectively improving the emergency response speed and management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flowchart of a method for identifying foreign object intrusion in a straight-line section of urban rail transit according to the present invention;

[0019] Figure 2 is a block diagram of a foreign object intrusion monitoring system for a straight-line section of urban rail transit according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0020] The following describes the specific embodiments of the present invention with reference to the drawings, so that those skilled in the art can better understand the present invention. It should be particularly noted that in the following description, when the detailed description of known functions and designs may dilute the main content of the present invention, these descriptions will be omitted here.

[0021] Embodiment

[0022] Figure 1 is a flowchart of a method for identifying foreign object intrusion in a straight-line section of urban rail transit according to the present invention.

[0023] In this embodiment, as Figure 1 shown, a method for identifying foreign object intrusion in a straight-line section of urban rail transit according to the present invention includes the following steps:

[0024] (1) Set up a foreign object intrusion detection center system with a straight section at the urban rail transit line control center. Set up a dedicated interface for train position information between the foreign object intrusion detection center system and the train control system to obtain the train position information K in real time;

[0025] (1.1) The existing train control system of urban rail transit contains train real-time position information. Obtain the train position information by setting up a dedicated interface for train position information at the rail transit line control center, and transmit the train position information into the foreign object intrusion detection center system to obtain the corresponding train real-time position system K.

[0026] (1.2) Create a central system line map according to the mileage characteristics of the rail transit line, which can map the train's traveling position into the line map and establish a real-time display of the train's position.

[0027] (2) Set up an image detection device and a radar detection device at the end of the locomotive to capture real-time vision images in the train's traveling direction and detect real-time radar data in the train's traveling direction on the straight section;

[0028] In this embodiment, the image detection device uses a vision sensor, and the radar detection device uses a millimeter-wave radar; the acquisition process of real-time vision images and real-time radar data is as follows:

[0029] (2.1) Set the image detection device at the center position of the locomotive end, and use the image detection device to capture real-time vision images in the train's traveling direction on the straight section;

[0030] (2.2) Set the radar detection device above the image detection device so that the radar detection device is located on the central axis of the track line; then, with the radar detection device as the origin, the train's traveling direction as the longitudinal coordinate axis direction, and the direction perpendicular to the track line as the transverse coordinate axis direction, construct a rectangular coordinate system;

[0031] In this embodiment, detecting an object means that the radar signal continuously detects a certain object exceeding a preset detection threshold. If the number of times of detecting the object is less than the preset detection threshold, it is considered that the object is lost or the detection fails, and the process of processing the detection data in the next step is not entered. Only when the number of times of detecting the object exceeds the preset detection threshold, the process of processing the detection data in the next step is entered. In this embodiment, the radar detection device uses a triangular wave modulation signal for foreign object detection.

[0032] (2.3) According to the difference frequency of the Doppler frequency shift, use the radar detection device to detect the distance, angle, and relative speed of the object in the train's traveling direction until the object is detected;

[0033] (2.4) Number the detected objects, and combine the numbers of the objects with the corresponding distances, angles, and relative speeds of the objects to obtain real-time radar data in the train traveling direction.

[0034] (3) Use the foreign object detection neural network in the train on-vehicle computing device to identify and detect the real-time field of view image, and obtain the field of view image feature detection result;

[0035] (3.1) Obtain a number of field of view images with or without foreign objects, and perform weighted average grayscale processing and Gaussian filtering on the field of view images in sequence to obtain a preprocessed field of view image;

[0036] In this embodiment, the model of weighted average grayscale processing is:

[0037] Gray(a,b) = (ω R R(a,b) + ω G G(a,b) + ω B B(a,b))

[0038] Among them, Gray(a,b) represents the grayscale value of the pixel at position (a,b) in the field of view image, ω R , ω G , ω B respectively represent the weight coefficients of the red, green, and blue channels, R(a,b), G(a,b), B(a,b) respectively represent the red, green, and blue channel pixel values of the pixel at position (a,b) in the field of view image, a is the abscissa of the field of view image, and b is the ordinate of the field of view image;

[0039] In this embodiment, the model of Gaussian filtering processing is:

[0040]

[0041] Among them, g(a,b) represents the Gaussian filtering result of the pixel at position (a,b) in the field of view grayscale image, e represents the exponential base constant, and σ represents the standard deviation of the Gaussian distribution;

[0042] (3.2) Perform histogram equalization processing and contrast enhancement processing on the preprocessed field of view image to obtain an enhanced field of view image;

[0043] (3.3) Use the sobel operator to perform edge detection on the enhanced field of view image to obtain the edge features of the field of view image;

[0044] In this embodiment, the gradient of the Gaussian-smoothed field of view image in the horizontal direction is calculated using a Sobel operator convolution kernel in the horizontal direction, and the gradient of the Gaussian-smoothed field of view image in the vertical direction is calculated using a Sobel operator convolution kernel in the vertical direction. Then, by calculating the magnitude of the gradients in the two directions, the edge features of the field of view image are obtained.

[0045] (3.4) According to the edge features of the field of view image, use the Hough transform to detect the straight lines of the image track to obtain the straight line track features of the field of view image;

[0046] In this embodiment, the method of Hough transform line detection is as follows: convert the field of view image to the polar coordinate parameter space and discretize it into a grid; according to the edge features of the field of view image, calculate all possible straight lines corresponding to the pixel points with edge features in the field of view image in the parameter space respectively, and perform cumulative counting at the parameter positions of the straight lines; find the pixel points whose cumulative counting results exceed the set threshold, and correspondingly obtain the straight line parameters in the field of view image, so as to obtain the track features of the field of view image.

[0047] (3.5) Set the detection limit for track foreign object intrusion;

[0048]

[0049] Among them, u ar (v i' ) represents the right position point of the detection limit for track foreign object intrusion, u ri' represents the characteristic point of the right track of the subway line, u li' represents the characteristic point of the left track of the subway line, u al (v i' ) represents the left position point of the detection limit for track foreign object detection, v i' represents the transverse coordinate of the i'-th track line;

[0050] (3.6) Use depthwise separable convolution to replace the standard convolution in the backbone network and feature fusion network of the traditional yolov11, and then adopt a top-down path aggregation mechanism and lateral connections to strengthen the image feature extraction and analysis between the layers of the feature fusion network, and use the C2f-Faster-EMA feature fusion strategy to fuse the image feature information of different scales to be transmitted to the neck network to obtain the improved yolov11 neural network;

[0051] In this embodiment, after replacing the standard convolutions in the backbone network and the feature fusion network of yolov11 with depthwise separable convolutions, since depthwise separable convolutions include depth convolutions and point convolutions, feature extraction is first performed on each input channel and then channel fusion is carried out, which can significantly reduce the number of parameters and computational complexity without affecting the accuracy. In addition, by adopting an automatic downward path aggregation mechanism and lateral connections, the additional computational overhead can be reduced. Using the C2f-Faster-EMA feature fusion strategy can better fuse information at different scales while controlling the computational overhead.

[0052] (3.7) Based on the track line foreign object intrusion dataset, and based on the straight track features of the field of view image and the detection limit of track foreign object intrusion, the improved yolov11 neural network is trained for foreign object detection and recognition to obtain a neural network for detecting foreign objects in the field of view image.

[0053] In this embodiment, the detection limit of track foreign object intrusion can efficiently determine the region of interest of the detection limit of track foreign object intrusion during the foreign object detection and recognition training for monitoring and recognition training. When training the improved yolov11 neural network, half-precision floating-point numbers can be used for gradient update to reduce the video memory consumption and improve the GPU utilization rate, thereby accelerating the training speed without affecting the convergence.

[0054] (3.8) Use the neural network for detecting foreign objects in the field of view image to detect the real-time field of view image to obtain the detection result of the field of view image features.

[0055] (4) According to the Kalman filtering algorithm, the on-train computing device of the train is used to identify and monitor the real-time radar data to obtain the distance Δx between the currently detected radar object and the locomotive end, and then the radar detection object distance Δx is uploaded to the foreign object intrusion detection center system in real time through the vehicle-ground wireless communication system.

[0056] (4.1) Set a continuous monitoring time window for the real-time radar data in the traveling direction of the train straight section, and count the number of times the object is successfully detected within the continuous monitoring time window as the object detection times.

[0057] (4.2) If the object detection times are less than the first detection times threshold, it is determined that the track line state is a foreign object intrusion disappearance state, and return to step (4.1); otherwise, enter step (4.3).

[0058] (4.3) If the object detection times are greater than the second detection times threshold, it is determined that the track line state is a foreign object intrusion continuous state, and enter step (4.4); otherwise, directly enter step (4.1).

[0059] (4.4) Construct a foreign object intrusion state space model based on the real-time radar data.

[0060]

[0061] v kτ = v k(τ-1) + a k(τ-1) ×Δt

[0062] where p kτ represents the position of the k-th object at time τ, v kτ represents the velocity of the k-th object at time τ, a k(τ-1) represents the acceleration of the k-th object at time τ - 1, and Δt represents the length of the adjacent time interval;

[0063] (4.5) Based on the foreign object intrusion state space model, construct the foreign object intrusion state equation;

[0064]

[0065] where ω τ-1 represents the process noise at time τ - 1, z τ represents the position measurement vector at time τ, H τ-1 represents the observation matrix at time τ - 1, and S τ-1 represents the measurement noise at time τ - 1;

[0066] (4.6) According to the Kalman filter algorithm, perform time update and state update on the foreign object intrusion state equation to obtain the distance Δx between the currently detected object by the radar and the end of the locomotive;

[0067] (4.7) Use the existing train-ground wireless communication system in urban rail transit to upload the radar-detected object distance Δx to the central system in real time.

[0068] In this embodiment, the train control system and the train-ground wireless communication system are both existing systems in urban rail transit.

[0069] (5) According to the data synchronization frequency, use the on-board computing device of the train to perform target position fusion on the current position of the radar-detected object and the detection result of the vision image features to obtain the foreign object intrusion detection result;

[0070] (5.1) Obtain the data synchronization frequency according to the detection frequency of the radar detection device and the detection frequency of the image detection device;

[0071] (5.2) Obtain the conversion model between the radar coordinate system and the world coordinate system, the conversion model between the pixel coordinate system and the camera coordinate system, and the conversion model between the camera coordinate system and the world coordinate system;

[0072] (5.3) Based on the conversion model between the radar coordinate system and the world coordinate system, convert the current position of the object detected by the radar into the current radar detection position in the world coordinate system;

[0073] (5.4) Based on the conversion model between the pixel coordinate system and the camera coordinate system and the conversion model between the camera coordinate system and the world coordinate system, convert the detection result of the field of view image features into the image detection result in the world coordinate system;

[0074] (5.5) According to the data synchronization frequency, perform target position fusion on the current radar detection position and the image detection result at the same moment based on the following target association model to obtain the vehicle-mounted intrusion detection result;

[0075] Among them, the target association model is:

[0076]

[0077] Among them, Δd represents the distance between the current radar detection position and the position of the center point of the object in the image detection result, O R represents the current radar detection position, O C represents the position of the center point of the object in the image detection result, l1 represents the first association degree threshold, IOU represents the intersection over union of the radar and the image detection object area, IOU represents the contour area of the radar detection object, S C represents the contour area of the image detection object, and l2 represents the second association degree threshold.

[0078] In this embodiment, when the distance between the current radar detection position and the position of the center point of the object in the image detection result is less than the first association degree threshold, and the intersection over union of the radar and the image detection object area is greater than the second association degree threshold, the object of the target position fusion can be effectively used as the vehicle-mounted intrusion detection result.

[0079] (6) The foreign object intrusion detection center system continuously sums the train position information K and the distance Δx of the object detected by the radar to calculate the foreign object position information, and then performs mapping and warning on the line map of the foreign object intrusion detection center system;

[0080] (6.1) Upload the vehicle-mounted intrusion detection result to the foreign object intrusion detection center system through the existing vehicle-ground wireless communication system of urban rail transit;

[0081] (6.2) The foreign object intrusion detection center system continuously sums the train position information K and the distance Δx of the object detected by the radar to calculate the foreign object position information K + Δx, and performs mapping and warning on the center system line map in combination with the foreign object intrusion detection result.

[0082] Such as Figure 2As shown in the figure, the present invention provides a detection system for identifying foreign object intrusion on the straight section of urban rail transit lines based on the above-mentioned method for identifying foreign object intrusion on the straight section of urban rail transit lines, that is, a detection system for foreign object intrusion on the straight section of urban rail transit lines; the detection system for foreign object intrusion on the straight section of urban rail transit lines includes: a foreign object intrusion detection center system set up at the urban rail transit line control center and a vehicle-mounted foreign object intrusion detection subsystem set up at the train locomotive.

[0083] The vehicle-mounted foreign object intrusion detection subsystem can detect and identify foreign object intrusion by combining images and radar, use the train on-board computing device to process the detection results of the intrusion foreign object and encapsulate them into upload information data packets, and then send the data packet information to the foreign object intrusion detection center system through the existing vehicle-ground wireless communication system of rail transit.

[0084] The foreign object intrusion detection center system creates a simulated line map according to the line mileage characteristics. There is a dedicated interface for train position information between the foreign object intrusion detection center system and the existing train control system, and the train position information can be obtained in real time. The foreign object intrusion detection center system parses the information data packets uploaded by the train on-board computing device in real time, and the position information of the intrusion foreign object can be obtained through summation calculation. By using the foreign object intrusion detection center system, the visual management of foreign object intrusion and train position mapping alarm on the line map is realized.

[0085] The present invention comprehensively applies technologies such as the integrated monitoring of vehicle-mounted vision and radar, neural network recognition, Kalman filtering algorithm and visual management, etc., comprehensively improves the detection efficiency of foreign object intrusion on the straight section of urban rail transit lines, and overall improves the real-time performance and accuracy of foreign object intrusion detection, the intelligent integration and target positioning ability, as well as the visual management and rapid response ability, providing a strong guarantee for the safe operation of trains.

[0086] Although the above-described illustrative specific embodiments of the present invention have been described to facilitate the understanding of the present invention by those skilled in the art, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions created using the concept of the present invention are within the scope of protection.

Claims

1. A method for identifying foreign objects intruding on a straight line segment of urban rail transit, characterized in that: The steps include: (1) A foreign body intrusion detection center system for straight sections is set up in the urban rail transit line control center, and a dedicated train position information interface is set up between the foreign body intrusion detection center system and the train control system to obtain the train position information K in real time; (2) Install an image detection device and a radar detection device at the end of the locomotive to capture a real-time field of view image of the train's traveling direction and to detect and obtain real-time radar data of the train's traveling direction in a straight section; (3) Using the visual field image foreign body detection neural network in the train onboard computing device to identify and detect the real-time visual field image, and obtain the visual field image feature detection result; (4) According to the Kalman filter algorithm, the train-mounted computing device is used to identify and monitor the real-time radar data to obtain the distance Δx between the current radar detection object and the end of the locomotive. Then, the train-to-ground wireless communication system is used to upload the radar detection object distance Δx to the foreign object intrusion detection center system in real time. (5) According to the data synchronization frequency, the train-mounted computing device is used to fuse the current radar detection object position and the field of view image feature detection results to obtain the foreign object intrusion detection result; (6) The foreign object intrusion detection center system creates a central system route map based on the mileage characteristics of the rail transit line, then maps the train's travel position to the route map, establishes a real-time train position display, and sums the train position information K with the radar detection object distance Δx in real time to calculate the foreign object position information, and then maps the alarm on the route map of the foreign object intrusion detection center system.

2. The identification method of the method for identifying the intrusion of foreign objects on the straight line of urban rail transit according to claim 1 is characterized in that: The step (1) comprises the following steps: (1.1) A dedicated interface for train location information is set up in the rail transit line control center to obtain the train location information from the train control system, and the train location information is transmitted to the foreign object intrusion detection center system to obtain the corresponding train real-time location system K; (1.2) Create a central system route map based on the mileage characteristics of the rail transit line, then map the train's travel position to the route map and establish a real-time train position display.

3. The method for identifying foreign objects intruding on a straight line segment of urban rail transit according to claim 1 is characterized in that: The step (2) comprises the following steps: (2.1) An image detection device is set at the center of the locomotive end, and the image detection device is used to capture a real-time field of view image of the train in the direction of travel of the straight segment; (2.2) Place the radar detection device above the image detection device so that the radar detection device is located on the central axis of the track line; then construct a rectangular coordinate system with the radar detection device as the origin, the train travel direction as the longitudinal coordinate axis, and the direction perpendicular to the track line as the transverse coordinate axis; (2.3) According to the difference frequency of the Doppler shift, the distance, angle and relative speed of the object in the direction of the train are detected by using a radar detection device until the object is detected; (2.4) The detected objects are numbered, and the object numbers are combined with the corresponding distance, angle and relative speed of the objects to obtain real-time radar data of the train's direction of travel.

4. The method for identifying foreign objects intruding on a straight line segment of urban rail transit according to claim 1 is characterized in that: The step (3) comprises the following steps: (3.1) Obtain a number of visual field images with or without foreign matter, and perform weighted average grayscale processing and Gaussian filtering processing on the visual field images in sequence to obtain pre-processed visual field images; (3.2) performing histogram equalization and contrast enhancement processing on the preprocessed visual field image to obtain an image-enhanced visual field image; (3.3) Use the Sobel operator to perform edge detection on the visual field image after image enhancement to obtain the edge features of the visual field image; (3.4) According to the edge features of the visual field image, the image track straight line detection is performed using Hough transform to obtain the visual field image straight line track features; (3.5) Set the limit for detecting foreign matter intrusion on the track; Among them, u ar (v i' ) represents the right side position point of the track foreign body intrusion detection limit, u ri' Indicates the characteristic point of the right track of the subway line, u li' Indicates the characteristic point of the left track of the subway line, u al (v i' ) represents the left side position point of the track foreign body detection limit, v i' represents the lateral coordinate of the i'th track line; (3.6) Use depthwise separable convolution to replace the standard convolution in the backbone network and feature fusion network of the traditional YOLOv11, and then use the top-down path aggregation mechanism and lateral connection to strengthen the image feature extraction and analysis between each layer of the feature fusion network, and use the C2f-Faster-EMA feature fusion strategy to fuse image feature information of different scales to transmit to the neck network to obtain the improved YOLOv11 neural network; (3.7) According to the foreign body intrusion limit dataset of urban rail transit track lines, based on the straight track features of the field of view image and the foreign body intrusion limit detection limit of the track, the improved yolov11 neural network is trained for foreign body detection and recognition, and the field of view image foreign body detection neural network is obtained; (3.8) Use the visual field image foreign body detection neural network to detect the real-time visual field image and obtain the visual field image feature detection result.

5. The method for identifying foreign objects intruding on a straight line segment of urban rail transit according to claim 1 is characterized in that: The step (4) comprises the following steps: (4.1) Setting a continuous monitoring time window for the real-time radar data in the straight section of the train, and counting the number of successful object detections within the continuous monitoring time window as the number of object detections; (4.2) If the object detection times are less than the first detection times threshold, the track line state is determined to be a foreign object invasion disappearance state, and the process returns to step (4.1); otherwise, the process proceeds to step (4.3); (4.3) If the number of object detections is greater than the second detection number threshold, the track line state is determined to be a foreign object intrusion continuous state, and the process goes to step (4.4); otherwise, the process goes directly to step (4.1); (4.4) Construct a foreign body intrusion state space model based on real-time radar data; in kτ =in k(τ-1) +a k(τ-1) ×Δt Among them, p kτ represents the position of the kth object at time τ, v kτ represents the velocity of the kth object at time τ, a k(τ-1) represents the acceleration of the kth object at time τ-1, and Δt represents the length of the adjacent time interval; (4.5) Based on the foreign body invasion state space model, construct the foreign body invasion state equation; Among them, ω τ-1 represents the process noise at time τ-1, z τ represents the position measurement vector at time τ, H τ-1 represents the observation matrix at time τ-1, S τ-1 represents the measurement noise at time τ-1; (4.6) According to the Kalman filter algorithm, the foreign body intrusion state equation is updated in time and state to obtain the distance Δx between the current radar detection object and the locomotive end; (4.7) Use the existing vehicle-to-ground wireless communication system of urban rail transit to upload the radar detection object distance Δx to the central system in real time.

6. The method for identifying foreign objects intruding on a straight line segment of urban rail transit according to claim 1, characterized in that: The step (5) comprises the following steps: (5.1) Obtaining a data synchronization frequency according to the detection frequency of the radar detection device and the detection frequency of the image detection device; (5.2) Obtain the conversion model between the radar coordinate system and the world coordinate system, the conversion model between the pixel coordinate system and the camera coordinate system, and the conversion model between the camera coordinate system and the world coordinate system; (5.3) Based on the conversion model between the radar coordinate system and the world coordinate system, the current radar detection object position is converted into the current radar detection position in the world coordinate system; (5.4) Based on the conversion model between the pixel coordinate system and the camera coordinate system and the conversion model between the camera coordinate system and the world coordinate system, the field of view image feature detection results are converted into image detection results in the world coordinate system; (5.5) According to the data synchronization frequency, the current radar detection position and image detection results at the same time are fused based on the following target association model to obtain the vehicle-mounted intrusion detection result; Among them, the target association model is: Wherein, Δd represents the distance between the current radar detection position and the center point of the object in the image detection result, O R Indicates the current radar detection position, O C represents the center point position of the object in the image detection result, l1 represents the first correlation threshold, IOU represents the intersection-over-union ratio of the radar and image detection object areas, IOU represents the contour area of ​​the radar detection object, S C represents the contour area of ​​the object detected in the image, and l2 represents the second association threshold.

7. The method for identifying foreign objects intruding on a straight line segment of urban rail transit according to claim 1, characterized in that: The step (6) comprises the following steps: (6.1) Upload the vehicle-borne intrusion detection results to the foreign object intrusion center system through the existing vehicle-to-ground wireless communication system of urban rail transit; (6.2) The foreign object intrusion center system sums the train position information K and the radar detection object distance Δx in real time, calculates the foreign object position information K+Δx, and maps the alarm on the center system route map based on the foreign object intrusion detection result.

Images