False alarm prevention method, system and equipment for obstacles in front of train, medium and product
By combining deep learning and railway-specific maps, and utilizing BeiDou satellite-based differential positioning and inertial navigation systems, the system achieves accurate detection of obstacles in front of trains, solving the problem of false alarms on curved railway sections and improving the reliability and accuracy of detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 西北铁道电子股份有限公司
- Filing Date
- 2026-01-20
- Publication Date
- 2026-04-28
AI Technical Summary
Existing technologies struggle to distinguish between visually overlapping fixed targets and real obstacles on curved railway sections, resulting in a high false alarm rate and a lack of three-dimensional depth understanding.
By combining deep learning algorithms and railway-specific maps, and by acquiring the real-time position of trains and image recognition, the BeiDou satellite-based differential positioning and inertial navigation system are used to determine the target offset and safety limits, thereby achieving accurate detection of both fixed and dynamic targets.
It reduces the false alarm rate, improves the reliability of obstacle detection, adapts to complex line environments, and ensures continuous monitoring across the entire line.
Smart Images
Figure CN121937979A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of obstacle recognition, and in particular to a method, system, device, medium and product for preventing false alarms about obstacles in front of a train. Background Technology
[0002] Currently, obstacle recognition technology on railway lines based solely on images relies primarily on the analysis and processing of two-dimensional planar images. This approach has an inherent technical bottleneck: on curved sections of railway, due to perspective projection, two parallel rails converge and project as a single line segment during imaging. Simultaneously, fixed equipment or structures located beside the track (such as signal lights, overhead contact line supports, mileage markers, etc.) and other trackside objects extending diagonally in front of the railway line in the image will visually overlap with the railway line during imaging due to specific viewing angles. This kind of "overlap" caused by perspective is extremely similar to real obstacles on the line in two-dimensional image features. Lacking a three-dimensional depth understanding of the scene, existing technologies struggle to effectively distinguish between this "visual overlap" and "physical encroachment" based solely on pixel information, easily misjudging harmless trackside fixed targets as foreign objects on the line, ultimately triggering unnecessary false alarms.
[0003] Therefore, based on the above problems, there is an urgent need to provide a method for preventing false alarms of obstacles in front of trains, so as to reduce the false alarm rate and improve the reliability of obstacle detection. Summary of the Invention
[0004] The purpose of this application is to provide a method, system, device, medium, and product for preventing false alarms of obstacles ahead of trains, so as to reduce the false alarm rate and improve the reliability of obstacle detection.
[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for preventing false alarms about obstacles ahead of a train, including: Obtain the real-time location of the train; Images are acquired within a preset range of the train track, and targets in the images are identified based on a deep learning algorithm to obtain target information; the target information includes: target coordinates and target category; the target category includes: fixed targets and dynamic targets; The target offset is determined based on the real-time position of the train and the target coordinates; the target offset is the distance between the target and the centerline of the train track. Based on the railway-specific map, determine the safety clearance between the train and the target, and the registered coordinates of the fixed target in the railway-specific map within the preset range of the train track; Based on the target information, target offset, safety clearance, and registered coordinates, an alarm signal is determined; the alarm signal includes: warning and alarm.
[0006] Optionally, obtaining the real-time location of the train specifically includes: The initial latitude and longitude coordinates of the train were determined using the BeiDou satellite-based differential positioning method. The continuous positioning coordinates of the train are determined using an inertial navigation system; Based on the initial latitude and longitude coordinates and the continuous positioning coordinates, the real-time position of the train is determined using the Kalman filter algorithm.
[0007] Optionally, the step of acquiring images within a preset range of the train track and identifying targets in the images based on a deep learning algorithm to obtain target information specifically includes: Images are collected within a preset range of the train track, and targets in the images are identified based on deep learning algorithms to obtain the target category and the pixel coordinates of the target. Based on the pixel coordinates, the target coordinates are determined using a projection transformation model.
[0008] Optionally, the railway-specific map includes: pre-stored registered coordinates of all fixed equipment along the line, registered coordinates of buildings, registered coordinates of legal obstacles, train line mileage, and safety clearance.
[0009] Optionally, determining the alarm signal based on target information, target offset, safety clearance, and registered coordinates specifically includes: When the target category is a fixed target, it is determined whether the target coordinates match the corresponding registered coordinates; if they match and the target offset is greater than or equal to the safety limit, no alarm signal is generated; if they do not match, an early warning is issued and it is determined whether the target offset is greater than or equal to the safety limit; if yes, no alarm signal is generated; if no, an alarm is issued. When the target category is a dynamic target, an early warning is issued and it is determined whether the target offset is greater than or equal to the safety limit; if yes, no alarm signal is generated; otherwise, an alarm is issued.
[0010] Optionally, when the target category is a fixed target, determining whether the target coordinates match the corresponding registered coordinates specifically includes: Using formula Determine whether the target coordinates match the corresponding registered coordinates; in, The x-coordinate of the target coordinates The x-coordinate of the registered coordinates. This is the error tolerance.
[0011] Secondly, this application provides a train obstacle warning system, including: The positioning module is used to obtain the real-time location of the train; The image recognition module is used to acquire images within a preset range of the train track and to identify targets in the images based on a deep learning algorithm to obtain target information. The target information includes: target coordinates and target category; the target category includes: fixed targets and dynamic targets. The target offset determination module is used to determine the target offset based on the real-time position of the train and the target coordinates; the target offset is the distance between the target and the centerline of the train track. The railway-specific map module is used to determine the safety clearance between the train and the target, and the registered coordinates of fixed targets in the railway-specific map within the preset range of the train track, based on the railway-specific map. The alarm signal determination module is used to determine the alarm signal based on the target information, target offset, safety clearance, and registered coordinates; the alarm signal includes: warning and alarm.
[0012] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for preventing false alarms of obstacles ahead of a train.
[0013] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for preventing false alarms of obstacles ahead of a train.
[0014] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for preventing false alarms of obstacles ahead of a train.
[0015] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, system, device, medium, and product for preventing false alarms about obstacles ahead of trains. By acquiring the real-time position of the train, continuous monitoring is ensured across the entire railway line. By combining the real-time position of the train with target recognition, a dual verification mechanism of position and image is established, enabling accurate detection of both fixed and dynamic targets. The introduction of a dedicated railway map to determine the registered coordinates of fixed targets avoids false alarms for legitimate fixed targets, reducing the false alarm rate. Based on target information, target offset, safety clearance, and registered coordinates, an alarm signal is determined, and foreign objects within a preset range on the train track are judged from multiple angles, further reducing the false alarm rate. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a flowchart illustrating a method for preventing false alarms about obstacles ahead of a train, according to one embodiment of this application. Detailed Implementation
[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0019] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0020] In one exemplary embodiment, such as Figure 1 As shown, a method for preventing false alarms about obstacles ahead of a train is provided, including the following S1 to S5. Wherein: S1: Obtain the real-time location of the train.
[0021] S1 specifically includes: S11: Determine the initial latitude and longitude coordinates of the train using the BeiDou satellite differential positioning method.
[0022] Based on satellite positioning signals and satellite-based differential station correction data, the initial latitude and longitude coordinates of the train are calculated using the BeiDou satellite-based differential positioning method. Specifically, reference points with known coordinates along the railway line (such as signal lights and bridge markers) are used as references. When the train passes these reference points, the actual coordinates (initial latitude and longitude coordinates) of the train are determined using the BeiDou satellite-based differential positioning method.
[0023] S12: Use an inertial navigation system to determine the continuous positioning coordinates of the train.
[0024] Inertial navigation systems can collect data such as the train's acceleration and angular velocity in real time, and further determine the train's continuous positioning coordinates.
[0025] S13: Based on the initial latitude and longitude coordinates and continuous positioning coordinates, determine the real-time position of the train using the Kalman filter algorithm.
[0026] The initial latitude and longitude coordinates and continuous positioning coordinates of the train at the same moment are compared to obtain the error between the two. The Kalman filter algorithm is used to feed the error back to the inertial navigation system to correct the continuous positioning coordinates of the train, and output high-frequency (frequency ≥ 100Hz) continuous positioning results and the train's line mileage information.
[0027] During train operation, the real-time position of the train is determined with a period of T seconds (e.g., T=0.1). ,in, Longitude As a dimension, The distance is the line mileage. This application combines the BeiDou satellite-based differential positioning method with an inertial navigation system to obtain the real-time position of the train with a positioning accuracy of less than 1 meter, and realizes dynamic calibration of the train positioning results. When the BeiDou signal disappears, this application can rely on the inertial navigation system and historical geographic data for continuous monitoring, that is, this application supports offline processing and has edge computing capabilities.
[0028] S2: Collect images within a preset range of the train track, and identify targets in the images based on a deep learning algorithm to obtain target information; target information includes: target coordinates and target category.
[0029] S2 specifically includes: S21: Collect images within a preset range of the train track, and identify targets in the images based on deep learning algorithms to obtain target categories and pixel coordinates.
[0030] During train operation, onboard cameras capture real-time images of a preset area along the train track, with the preset area defined by the train's real-time position. Centered on the track, within a radius R (R=5m), deep learning algorithms (such as YOLO and Faster R-CNN) are used to identify targets within this predetermined area, outputting the target category and pixel coordinates. The deep learning algorithm can initially distinguish target categories; specifically, target categories include fixed targets and dynamic targets. Fixed targets include fixed equipment, buildings, and suspected obstacles, while dynamic targets include pedestrians and animals. After initially obtaining the target categories, dynamic targets are marked as key monitoring objects.
[0031] S22: Determine the target coordinates based on the pixel coordinates and the projection transformation model.
[0032] A projection transformation model is established based on the installation parameters (height, pitch angle, and azimuth angle) of the vehicle-mounted camera, and the pixel coordinates of each target are then determined using the projection transformation model. Convert to geographic coordinates Also known as target coordinates, it establishes a spatial location mapping of the target; it also determines the height of each target. To determine whether the target is a ground intruder, among which Number the target.
[0033] The BeiDou-based differential positioning method proposed in this application can achieve real-time clock synchronization, ensuring the consistency between the real-time position of the train and the timestamp of the image.
[0034] S3: Determine the target offset based on the real-time position of the train and the target coordinates.
[0035] target offset This is the distance between the target and the centerline of the train track, determined by the target coordinates and the real-time position of the train.
[0036] S4: Based on the railway-specific map, determine the safety clearance between the train and the target, and the registered coordinates of the fixed target in the railway-specific map within the preset range of the train track.
[0037] The railway-specific map includes pre-stored coordinates of all fixed equipment along the line (such as signals, catenary supports, and mileage markers), coordinates of structures (such as bridges and tunnel entrances), coordinates of legal obstacles (such as guardrails) (latitude and longitude coordinates), train line mileage, and safety clearance. (Minimum safe distance from the centerline of the track); in addition, railway-specific maps also include the spatial dimensions (length, width, and height) of fixed equipment, buildings, and legal obstacles.
[0038] Search for real-time train locations on railway-specific maps. Centered on a fixed target within a radius R (R=5m), the registered coordinates of the fixed target in the railway-specific map within the preset range of the train track are further obtained. .
[0039] S5: Determine the alarm signal based on the target information, target offset, safety clearance, and registered coordinates; the alarm signal includes: warning and alarm.
[0040] When the target category is a fixed target, it is determined whether the target coordinates match the corresponding registered coordinates. If they match and the target offset is greater than or equal to the safety clearance, it is determined to be a legal fixed target, and no alarm signal is generated. If they do not match, it is determined to be a suspected intrusive obstacle, an early warning is issued, and it is determined whether the target offset is greater than or equal to the safety clearance (determining whether the following conditions are met). If yes, no alarm signal is generated; otherwise, an alarm is triggered. Furthermore, in specific application scenarios, there is no situation where the target coordinates match the corresponding registered coordinates but the target offset is less than the safety limit; therefore, this application does not consider such cases.
[0041] When the target category is dynamic, an early warning is issued and it is determined whether the target offset is greater than or equal to the safety clearance (to determine whether the safety clearance is met). If yes, no alarm signal is generated; otherwise, an alarm is triggered.
[0042] Specifically, the formula for determining whether the target coordinates match the corresponding registered coordinates is as follows: ; in, The x-coordinate of the target coordinates The x-coordinate of the registered coordinates. This is the error tolerance.
[0043] In an exemplary embodiment, when the target category is a dynamic target and the target offset is less than the safety clearance (satisfying...) When the target is determined to be a suspected obstacle, an early warning is triggered and a secondary verification is initiated (for example, if a dynamic target is detected for N consecutive frames and its position continues to approach), an alarm signal is generated based on the results of the secondary verification.
[0044] The received alarm signal, along with the target location, target category, and hazard level, is sent to the train control system, while a prompt message (such as "fixed equipment, safe") is sent to the train driver.
[0045] This application introduces high-precision real-time train location data and a dedicated railway map to match the image recognition results of fixed targets within a preset track area with their registered coordinates. This eliminates false alarms caused by overlapping images due to line-of-sight angles (such as fixed equipment whose rail projections overlap on curved sections), solving the problem that existing pure image recognition technologies cannot distinguish between fixed targets and foreign objects, and significantly reducing the false alarm rate. Furthermore, this application integrates positioning data and image recognition results to establish a dual "position-image" verification mechanism, enabling accurate detection of dynamic targets (such as pedestrians suddenly entering) and unregistered fixed targets (such as falling rocks and fallen trees). This avoids false alarms for legitimate fixed targets within the dedicated railway map, improving the reliability of obstacle detection. In addition, this application utilizes an inertial navigation system to compensate for positioning interruptions in scenarios where BeiDou signals are obstructed (such as tunnels and mountainous areas), combined with the offline matching function of the dedicated railway map, ensuring continuous monitoring across the entire railway line. This adapts to complex track environments and enhances system robustness.
[0046] Based on the same inventive concept, this application also provides a train obstacle forward false alarm prevention system. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more train obstacle forward false alarm prevention system embodiments provided below can be found in the limitations of the train obstacle forward false alarm prevention method described above, and will not be repeated here.
[0047] In one exemplary embodiment, a train obstacle warning system is provided, comprising: The positioning module is used to obtain the real-time location of the train; The image recognition module is used to acquire images within a preset range of the train track and to identify targets in the images based on a deep learning algorithm to obtain target information. The target information includes: target coordinates and target category; the target category includes: fixed targets and dynamic targets. The target offset determination module is used to determine the target offset based on the real-time position of the train and the target coordinates; the target offset is the distance between the target and the centerline of the train track. The railway-specific map module is used to determine the safety clearance between the train and the target, and the registered coordinates of fixed targets in the railway-specific map within the preset range of the train track, based on the railway-specific map. The alarm signal determination module is used to determine the alarm signal based on the target information, target offset, safety clearance, and registered coordinates; the alarm signal includes: warning and alarm.
[0048] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores false alarm data for obstacles ahead of the train. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for preventing false alarms for obstacles ahead of the train.
[0049] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0050] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0051] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0052] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0053] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0054] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0055] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0056] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method for preventing false alarms about obstacles ahead of a train, characterized in that, The method for preventing false alarms about obstacles ahead of the train includes: Obtain the real-time location of the train; Images are acquired within a preset range of the train track, and targets in the images are identified based on a deep learning algorithm to obtain target information; the target information includes: target coordinates and target category; the target category includes: fixed targets and dynamic targets; The target offset is determined based on the real-time position of the train and the target coordinates; the target offset is the distance between the target and the centerline of the train track. Based on the railway-specific map, determine the safety clearance between the train and the target, and the registered coordinates of the fixed target in the railway-specific map within the preset range of the train track; Based on the target information, target offset, safety clearance, and registered coordinates, an alarm signal is determined; the alarm signal includes: warning and alarm.
2. The method for preventing false alarms about obstacles ahead of a train according to claim 1, characterized in that, The acquisition of the train's real-time location specifically includes: The initial latitude and longitude coordinates of the train were determined using the BeiDou satellite-based differential positioning method. The continuous positioning coordinates of the train are determined using an inertial navigation system; Based on the initial latitude and longitude coordinates and the continuous positioning coordinates, the real-time position of the train is determined using the Kalman filter algorithm.
3. The method for preventing false alarms about obstacles ahead of a train according to claim 1, characterized in that, The process involves acquiring images within a preset range of the train track and identifying targets in the images using a deep learning algorithm to obtain target information. Specifically, this includes: Images are collected within a preset range of the train track, and targets in the images are identified based on deep learning algorithms to obtain the target category and the pixel coordinates of the target. Based on the pixel coordinates, the target coordinates are determined using a projection transformation model.
4. The method for preventing false alarms about obstacles ahead of a train according to claim 1, characterized in that, The railway-specific map includes: pre-stored registered coordinates of all fixed equipment along the line, registered coordinates of buildings, registered coordinates of legal obstacles, train line mileage, and safety clearance.
5. The method for preventing false alarms about obstacles ahead of a train according to claim 1, characterized in that, The process of determining the alarm signal based on target information, target offset, safety clearance, and registered coordinates specifically includes: When the target category is a fixed target, it is determined whether the target coordinates match the corresponding registered coordinates; if they match and the target offset is greater than or equal to the safety limit, no alarm signal is generated; if they do not match, an early warning is issued and it is determined whether the target offset is greater than or equal to the safety limit; if yes, no alarm signal is generated; if no, an alarm is issued. When the target category is a dynamic target, an early warning is issued and it is determined whether the target offset is greater than or equal to the safety limit; if yes, no alarm signal is generated; otherwise, an alarm is issued.
6. The method for preventing false alarms about obstacles ahead of a train according to claim 5, characterized in that, When the target category is a fixed target, determining whether the target coordinates match the corresponding registered coordinates specifically includes: Using formula Determine whether the target coordinates match the corresponding registered coordinates; in, The x-coordinate of the target coordinates. The x-coordinate of the registered coordinates. This is the tolerance for error.
7. A train obstacle warning system, characterized in that, The train obstacle forward false alarm prevention system includes: The positioning module is used to obtain the real-time location of the train; The image recognition module is used to acquire images within a preset range of the train track and identify targets in the images based on a deep learning algorithm to obtain target information; the target information includes: target coordinates and target category; the target category includes: fixed targets and dynamic targets; The target offset determination module is used to determine the target offset based on the real-time position of the train and the target coordinates; the target offset is the distance between the target and the centerline of the train track. The railway-specific map module is used to determine the safety clearance between the train and the target, and the registered coordinates of fixed targets in the railway-specific map within the preset range of the train track, based on the railway-specific map. The alarm signal determination module is used to determine the alarm signal based on the target information, target offset, safety clearance, and registered coordinates; the alarm signal includes: warning and alarm.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the train obstacle forward false alarm method according to any one of claims 1-6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for preventing false alarms of obstacles ahead of the train as described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for preventing false alarms of obstacles ahead of the train as described in any one of claims 1-6.