Train track detection method and system based on laser radar

Abstract

The invention relates to the technical field of image detection, and discloses a train track detection method and system based on a laser radar, and the method comprises the steps: collecting real-time monitoring data of the laser radar, and judging whether to start image detection or not according to the real-time monitoring data and radio frequency identification information; when it is judged that the image detection is started, turnout switching action image data are collected, image data after turnout switching are determined, the image data after turnout switching are processed based on Canny edge detection, and image data at the track connection position are obtained; according to the image data at the track joint, determining a straight line angle after turnout switching and a track gauge after turnout switching, and judging whether turnout switching is abnormal or not; when it is judged that abnormity exists, the deviation value is determined according to the straight line angle after turnout switching and the track gauge after turnout switching, whether the deviation value is adjusted or not is determined according to the image data of the track connection position, and the final deviation value is determined; the method can adapt to a complex track environment, and the automation level of track detection is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of image detection technology, and in particular to a laser radar-based train track detection method and system. Background Art

[0002] In modern rail transit systems, the status of train tracks is directly related to driving safety, especially during the track switch switching (track changing) process. The track connection accuracy and whether the track gauge meets the standards will affect the stable operation of the train.

[0003] Existing track inspection technologies primarily rely on fixed sensors or manual inspections. Manual inspections, however, are limited by high labor costs and low inspection frequency, making them inadequate for real-time monitoring of densely populated tracks. Even with the use of sensor technology, while fixed sensors (such as track circuits and pressure sensors) can provide some track status information, they still suffer from limited detection accuracy and delayed data updates when dealing with complex switch switching environments. Furthermore, current analysis of switch switching status typically relies on a single detection method, such as laser ranging, and lacks the ability to identify and adjust anomalies based on the fusion of multiple parameters (such as gauge and angle), making it difficult to accurately determine anomalies in switch switching scenarios.

[0004] Therefore, it is necessary to design a train track detection method and system based on lidar to solve the problems existing in current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a train track detection method and system based on laser radar, aiming to solve the problems of low detection efficiency and poor reliability of detection results in the current train track switch switching detection.

[0006] In one aspect, the present invention provides a train track detection method based on laser radar, comprising: Collecting real-time monitoring data from a laser radar, and determining whether to enable image detection based on the real-time monitoring data and radio frequency identification (RFI) identification information; When it is determined that the image detection is turned on, the image data of the switch switching action is collected, the image data after the switch switching is determined, and the image data after the switch switching is processed based on Canny edge detection to obtain the image data of the track connection; Determine the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and judge whether there is a switch switching abnormality; When it is determined that an abnormality exists, the deviation value is determined based on the straight line angle after the switch switching and the track gauge after the switch switching, and whether the deviation value is adjusted is determined based on the image data of the track connection to determine the final deviation value.

[0007] Furthermore, the determining whether to enable image detection based on the real-time monitoring data and the radio frequency identification identification information includes: Comparing the real-time monitoring data with a minimum data threshold, and determining to start the image detection when the real-time monitoring data is less than the minimum data threshold and the radio frequency identification identification information is consistent with the train preset information; When the real-time monitoring data is greater than or equal to the minimum data threshold, or when the radio frequency identification identification information is inconsistent with the train preset information, it is determined that the image detection is not enabled.

[0008] Furthermore, the image data after the switch switching is processed based on Canny edge detection to obtain the image data of the track connection, including: Performing Gaussian filtering on the image data after the switch switching, and adjusting the standard deviation of the Gaussian kernel to smooth the image; Calculating the gradient amplitude and direction of the image data after the switch switching, calculating the gradients in the horizontal and vertical directions based on the Sobel operator, synthesizing the total gradient G according to the gradients in the horizontal and vertical directions, and obtaining the gradient direction; Based on non-maximum suppression, adjacent pixels are checked according to the gradient direction, and only pixels with the local maximum gradient value are retained to remove redundant information; Based on double critical value detection, the edge strength is classified to determine the points to be retained and removed; Edge connection is performed according to the reserved points, and possible edges are tracked. If the possible edge is connected to a strong edge, it is retained; otherwise, it is removed, and finally the image data of the track connection is obtained.

[0009] Furthermore, based on the dual critical value detection, the edge strength is classified to determine the points to be retained and removed, including: A high critical value Th and a low critical value Tl are set. When the total gradient G of an edge pixel is greater than the high critical value Th, the edge pixel is defined as a strong edge and retained. When Th≥G≥Tl, the edge pixel is defined as a possible edge and retained. When the total gradient G of an edge pixel is less than the low critical value Tl, the edge pixel is defined as a weak edge and removed.

[0010] Furthermore, determining the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and judging whether there is a switch switching abnormality, includes: Marking all track positions according to the track connection image data, and arbitrarily selecting track positions to obtain several track gauges, and obtaining the straight line angle after the switch is switched according to the track connection image data; Comparing the straight line angle with a preset switch switching angle, and comparing all the track gauges with the standard track gauge, and determining whether there is a switch switching anomaly based on the comparison results; When the straight line angle is equal to the preset switch switching angle, and the track gauges of all the tracks are equal to the standard track gauge, it is determined that there is no switch switching abnormality; When the straight line angle is not equal to the preset switch switching angle, or the track gauge is not equal to the standard track gauge, it is determined that a switch switching abnormality exists.

[0011] Furthermore, when determining the deviation value according to the straight line angle after the switch switching and the track gauge after the switch switching, it includes: Obtaining an angle deviation ratio based on the straight line angle after the switch switching and a preset switch switching angle. When obtaining the angle deviation ratio, calculating the absolute value of the difference between the straight line angle after the switch switching and the preset switch switching angle, obtaining a ratio of the absolute value of the difference to the preset switch switching angle, and using the angle ratio as the angle deviation ratio. Obtaining a track gauge deviation ratio based on the track gauge after the switch switching and the standard track gauge. When obtaining the track gauge deviation ratio, extracting the track gauge corresponding to the maximum difference between the track gauge and the standard track gauge, obtaining a ratio of the maximum difference to the standard track gauge, and using the track gauge ratio as the track gauge deviation ratio; Obtain a total deviation ratio based on the angle deviation ratio and the gauge deviation ratio, where the total deviation ratio is the sum of the angle deviation ratio and the gauge deviation ratio, compare the total deviation ratio with a first-level deviation ratio threshold and a second-level deviation ratio threshold, respectively, and determine a deviation value based on the comparison results, where the first-level deviation ratio threshold is less than the second-level deviation ratio threshold; When the total deviation ratio is less than or equal to the first-level deviation ratio threshold, the deviation value is determined to be a first-level deviation value; when the total deviation ratio is greater than the first-level deviation ratio threshold and less than or equal to the second-level deviation ratio threshold, the deviation value is determined to be a second-level deviation value; when the total deviation ratio is greater than the second-level deviation ratio threshold, the deviation value is determined to be a third-level deviation value; and the first-level deviation value is less than the second-level deviation value, and the second-level deviation value is less than the third-level deviation value.

[0012] Furthermore, determining whether to adjust the deviation value based on the image data of the track connection includes: Determine whether there is debris blocking the track joint according to the joint image data, and determine whether to adjust the deviation value according to the debris situation; When there are debris at the track joint, it is determined that the deviation value should be adjusted; When there is no debris at the track joint, it is determined that the deviation value is not adjusted, and the deviation value is used as the final deviation value.

[0013] Furthermore, determining whether to adjust the deviation value includes: Obtain debris size data, determine a correction coefficient based on the debris size data, adjust the deviation value to determine the final deviation value, the correction coefficient is proportional to the debris size data, and the final deviation value is the product of the deviation value and the correction coefficient, and the correction coefficient ranges from 1 to 1.5, where the correction coefficient includes 1.5 and excludes 1.

[0014] Compared with the existing technology, the beneficial effects of the present invention are: by intelligently triggering image detection through laser radar combined with radio frequency identification identification information, the redundant data problem caused by the traditional continuous acquisition mode is avoided, and the detection efficiency and system response speed are improved. During the switch switching process, Canny edge detection is used to process the image data after the switch switching, accurately extract the characteristic information of the track connection, and can capture the track boundary and gauge changes. Through the track straight angle and gauge after the switch switching, it is possible to effectively identify whether the track switching is abnormal, and improve the accuracy of switch switching status monitoring. When an anomaly is detected, the deviation value analysis is further combined with the image data of the track connection, and support adjustment based on parameter fusion to improve the accuracy of anomaly judgment, reduce the error rate, and thus enhance the safety and reliability of track switch switching.

[0015] On the other hand, the present application also provides a laser radar-based train track detection system for applying the above-mentioned laser radar-based train track detection method, comprising: A data acquisition module is configured to collect real-time monitoring data of the laser radar and determine whether to enable image detection based on the real-time monitoring data and radio frequency identification identification information; a data processing module configured to, when it is determined that the image detection is enabled, collect image data of the switch switching action, determine image data after the switch switching, process the image data after the switch switching based on Canny edge detection, and obtain image data of the track connection; a state determination module configured to determine the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and to determine whether there is a switch switching abnormality; The deviation correction module is configured to determine the deviation value according to the straight line angle after the switch switching and the track gauge after the switch switching when the state judgment module determines that an abnormality exists, and determine whether to adjust the deviation value according to the image data of the track connection to determine the final deviation value.

[0016] Furthermore, it also includes: an alarm module, configured to determine an alarm level according to the final deviation value, and the alarm level is proportional to the final deviation value.

[0017] It is understandable that the above-mentioned laser radar-based train track detection method and system have the same beneficial effects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings: Figure 1 A flowchart of a laser radar-based train track detection method provided in an embodiment of the present invention; Figure 2 This is a functional block diagram of a laser radar-based train track detection system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0020] In some embodiments of the present application, referring to FIG1 , a laser radar-based train track detection method includes: S100: Collecting real-time monitoring data of the laser radar, and determining whether to start image detection based on the real-time monitoring data and radio frequency identification identification information.

[0021] S200: When it is determined that image detection is turned on, image data of the switch switching action is collected, image data after the switch switching is determined, and the image data after the switch switching is processed based on Canny edge detection to obtain image data of the track connection.

[0022] S300: Determine the straight line angle after the switch and the track gauge after the switch according to the image data of the track connection point, and judge whether there is a switch switching abnormality.

[0023] S400: When it is determined that there is an abnormality, the deviation value is determined according to the straight line angle after the switch switching and the track gauge after the switch switching, and whether the deviation value is adjusted is determined according to the image data of the track connection to determine the final deviation value.

[0024] Specifically, S100 collects real-time lidar monitoring data, including the distance between the lidar and the train as it passes. The lidars are fixed on both sides of the track and use RFID tags to confirm the train's position. RFID tags can be placed on the train, and train information is identified by an RFID reader to determine whether the train has entered the switch switching area. When the lidar data matches the RFID tag information, indicating that the train is about to enter the switch switching area, a decision is made to initiate image detection, reducing resource consumption for unnecessary image acquisition and improving detection efficiency. When S200 determines that image detection is necessary, a camera is activated to capture the switch switching process and obtain track image data after the switch switching. To ensure data processing accuracy, the Canny edge detection algorithm is used to process the post-switch switching image data. Canny edge detection uses Gaussian filtering to remove noise and calculates the gradient change of the image to detect track boundaries. Combining non-maximum suppression and double threshold detection, it extracts edge features at the track junction, thereby obtaining clear image data of the track junction. After acquiring image data from the track junction in S300, the track's linear angle and track gauge after the switch is switched are determined based on edge detection results. The track's linear angle is detected using a Hough transform, and the track's inclination change is calculated to determine whether the track junction is successful. The track gauge is determined by measuring the pixel distance between the two track edges and converting it based on camera parameters to obtain the actual track width. If the linear angle or track gauge deviates from the set standard, it indicates a switch switching anomaly. Once a switch switching anomaly is detected in S400, a deviation value is further calculated, representing the actual track offset after the switch switching. This deviation value is calculated based on the image data from the track junction and combined with the track angle change and track gauge offset for error analysis. If the deviation value exceeds the safety threshold, it can be adjusted based on the actual situation at the track junction to reduce false positives. The final deviation value is determined to improve detection reliability.

[0025] It's understandable that leveraging LiDAR and RFID tag information to implement intelligent image detection triggering reduces the significant amount of invalid data collected by traditional methods and improves response efficiency. Canny edge detection achieves high-precision detection of track boundaries and determines the status of switch switching based on track angles and gauges, resolving the inability of traditional single-laser ranging methods to comprehensively analyze track conditions. After detecting an anomaly, the system adjusts the deviation value based on actual parameter analysis, improving the accuracy of switch switching anomaly detection and reducing the false positive rate, thereby enhancing the safety of switch switching.

[0026] In some embodiments of the present application, when determining whether to turn on image detection based on real-time monitoring data and radio frequency identification identification information, it includes: comparing the real-time monitoring data with the minimum data threshold; when the real-time monitoring data is less than the minimum data threshold and the radio frequency identification identification information is consistent with the train preset information, it is determined that image detection is turned on; when the real-time monitoring data is greater than or equal to the minimum data threshold, or the radio frequency identification identification information is inconsistent with the train preset information, it is determined that image detection is not turned on.

[0027] In some embodiments of the present application, processing the image data after the switch switching based on Canny edge detection to obtain the image data of the track connection includes: Perform Gaussian filtering on the image data after the switch is switched, and adjust the standard deviation of the Gaussian kernel to smooth the image; Calculate the gradient amplitude and direction of the image data after the switch is switched, calculate the horizontal and vertical gradients based on the Sobel operator, and synthesize the total gradient G based on the horizontal and vertical gradients to obtain the gradient direction; Based on non-maximum suppression, adjacent pixels are checked according to the gradient direction, and only pixels with the local maximum gradient value are retained to remove redundant information; Based on double critical value detection, the edge strength is classified to determine the points to be retained and removed; Edge connection is performed according to the retained points, and possible edges are tracked. If the possible edge is connected to the strong edge, it is retained, otherwise it is removed, and finally the image data of the track connection is obtained.

[0028] In some embodiments of the present application, based on dual threshold detection, edge strength is classified to determine whether to retain or remove points, including: Set a high critical value Th and a low critical value Tl. When the total gradient G of an edge pixel is greater than the high critical value Th, the edge pixel is defined as a strong edge and retained. When Th≥G≥Tl, the edge pixel is defined as a possible edge and retained. When the total gradient G of an edge pixel is less than the low critical value Tl, the edge pixel is defined as a weak edge and removed.

[0029] As can be seen, the use of Canny edge detection and refined processing makes track boundary extraction more accurate. In particular, the optimization of dual-threshold detection and edge connection ensures clear and continuous edges at track junctions, enabling more precise analysis of track status after switch switching compared to traditional methods. Furthermore, the use of multi-step optimized gradient calculation and non-maximum suppression reduces noise interference, improves track detection stability, and thus enhances the reliability of track switch switching state analysis.

[0030] In some embodiments of the present application, when determining the straight-line angle and the track gauge after the switch switching based on the image data of the track connection, and judging whether there is a switch switching abnormality, it includes: marking all track positions according to the image data of the track connection, and arbitrarily selecting track positions to obtain several track gauges, and obtaining the straight-line angle after the switch switching based on the image data of the track connection; comparing the straight-line angle with the preset switch switching angle, and comparing all track gauges with the standard track gauge, and judging whether there is a switch switching abnormality based on the comparison result; when the straight-line angle is equal to the preset switch switching angle, and all track gauges are equal to the standard track gauge, it is judged that there is no switch switching abnormality; when the straight-line angle is not equal to the preset switch switching angle, or there is a track gauge that is not equal to the standard track gauge, it is judged that there is a switch switching abnormality.

[0031] In some embodiments of the present application, when determining the deviation value based on the straight line angle after the switch switching and the track gauge after the switch switching, the method includes: obtaining the angle deviation ratio based on the straight line angle after the switch switching and the preset switch switching angle, calculating the absolute value of the difference between the straight line angle after the switch switching and the preset switch switching angle when obtaining the angle deviation ratio, obtaining the ratio of the absolute value of the difference to the preset switch switching angle, and using the angle ratio as the angle deviation ratio; obtaining the gauge deviation ratio based on the track gauge after the switch switching and the standard track gauge, extracting the track gauge corresponding to the maximum value of the difference between the track gauge and the standard track gauge when obtaining the gauge deviation ratio, obtaining the ratio of the maximum difference to the standard track gauge, and using the gauge ratio as the gauge deviation ratio; obtaining the total deviation ratio based on the angle deviation ratio and the gauge deviation ratio, the total deviation ratio being the sum of the angle deviation ratio and the gauge deviation ratio, comparing the total deviation ratio with the first-level deviation ratio threshold and the second-level deviation ratio threshold respectively, and determining the deviation value based on the comparison result, the first-level deviation ratio threshold being less than the second-level deviation ratio threshold; Specifically, when the total deviation ratio is less than or equal to the first-level deviation ratio threshold, the deviation value is determined to be the first-level deviation value; when the total deviation ratio is greater than the first-level deviation ratio threshold and less than or equal to the second-level deviation ratio threshold, the deviation value is determined to be the second-level deviation value; when the total deviation ratio is greater than the second-level deviation ratio threshold, the deviation value is determined to be the third-level deviation value; and the first-level deviation value is less than the second-level deviation value, and the second-level deviation value is less than the third-level deviation value.

[0032] As can be seen, calculating the combined percentage of angle deviation and gauge deviation effectively improves the accuracy of detecting anomalies in track switch switching. Compared to traditional single-parameter detection methods (such as monitoring gauge deviation alone), the fusion of angle deviation and gauge deviation percentages allows for a more comprehensive assessment of the severity of anomalies in track switch switching. The multi-level anomaly classification enables refined adjustments and early warnings, ensuring the safety and stability of train operations during track switch switching.

[0033] In some embodiments of the present application, determining whether to adjust the deviation value based on the image data at the track connection includes: Determine whether there are any debris blocking the track joint based on the image data of the joint, and determine whether to adjust the deviation value based on the debris situation; When there are debris at the track joint, the deviation value is adjusted; When there is no debris at the track joint, it is determined that the deviation value is not adjusted and the deviation value is used as the final deviation value.

[0034] In some embodiments of the present application, determining whether to adjust the deviation value includes: Obtain debris size data, determine a correction coefficient based on the debris size data, adjust the deviation value to determine a final deviation value, the correction coefficient is proportional to the debris size data, and the final deviation value is the product of the deviation value and the correction coefficient, and the correction coefficient ranges from 1 to 1.5, where the correction coefficient includes 1.5 and excludes 1.

[0035] Specifically, during turnout inspection, deviations are often caused by track angle deviation or gauge variation. However, these deviations may not be due to issues with the track itself, but rather to debris (such as gravel, snow, or other debris) temporarily covering the track edge. Using image data from the track joint, the system can determine if there is any obstruction and then decide whether to adjust the deviation.

[0036] It's understandable that the introduction of debris detection and size adjustment mechanisms effectively reduces misjudgments during track switch switching detection. Compared to traditional methods that determine deviations based solely on track angle and gauge, image analysis technology, combined with debris identification and correction factor calculation, enables more accurate deviation determination. The correction factor setting avoids error accumulation caused by overcorrection, ensuring the stability of detection results. This improves the reliability of track switch switching status detection.

[0037] In the above-mentioned embodiment, the use of LiDAR combined with radio frequency identification (RFID) identification information to intelligently trigger image detection avoids the redundant data problem caused by the traditional continuous acquisition mode, thereby improving detection efficiency and system response speed. During the switch switching process, Canny edge detection is used to process the image data after the switch switching, accurately extracting the characteristic information of the track connection, and capable of capturing track boundaries and track gauge changes. The track straight line angle and track gauge after the switch switching can effectively identify whether the track switching is abnormal, thereby improving the accuracy of switch switching status monitoring. When an anomaly is detected, the deviation value analysis is further combined with the image data of the track connection, and parameter fusion-based adjustments are supported to improve the accuracy of anomaly judgment and reduce the error rate, thereby enhancing the safety and reliability of track switch switching.

[0038] In another preferred embodiment based on the above embodiment, referring to FIG2 , this embodiment provides a laser radar-based train track detection system for applying the above laser radar-based train track detection method, including: A data acquisition module is configured to collect real-time monitoring data of the laser radar and determine whether to enable image detection based on the real-time monitoring data and radio frequency identification identification information; a data processing module configured to, when it is determined that the image detection is enabled, collect image data of the switch switching action, determine image data after the switch switching, process the image data after the switch switching based on Canny edge detection, and obtain image data of the track connection; a state determination module configured to determine the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and to determine whether there is a switch switching abnormality; The deviation correction module is configured to determine the deviation value according to the straight line angle after the switch switching and the track gauge after the switch switching when the state judgment module determines that an abnormality exists, and determine whether to adjust the deviation value according to the image data of the track connection to determine the final deviation value.

[0039] Furthermore, the alarm module is configured to determine an alarm level according to the final deviation value, and the alarm level is proportional to the final deviation value.

[0040] It's understandable that the intelligent triggering of image detection using LiDAR combined with RFID tag information avoids the redundant data issues associated with traditional continuous acquisition methods, improving detection efficiency and system response speed. During the switch switching process, Canny edge detection is used to process the image data after the switch switching, accurately extracting feature information at the track junction and capturing track boundary and gauge changes. The track straight line angle and gauge after the switch switching can effectively identify track switching anomalies, improving the accuracy of switch switching status monitoring. When an anomaly is detected, deviation value analysis is further performed in conjunction with the image data at the track junction, supporting parameter fusion-based adjustments to improve the accuracy of anomaly determination and reduce the false positive rate, thereby enhancing the safety and reliability of track switch switching.

[0041] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0042] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0043] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0044] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A train track detection method based on laser radar, characterized in that: include: Collecting real-time monitoring data from the laser radar, and determining whether to enable image detection based on the real-time monitoring data and radio frequency identification information; When it is determined that the image detection is turned on, the image data of the switch switching action is collected, the image data after the switch switching is determined, and the image data after the switch switching is processed based on Canny edge detection to obtain the image data of the track connection; Determine the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and judge whether there is a switch switching abnormality; When it is determined that an abnormality exists, the deviation value is determined based on the straight line angle after the switch switching and the track gauge after the switch switching, and whether the deviation value is adjusted is determined based on the image data of the track connection to determine the final deviation value.

2. The train track detection method based on laser radar according to claim 1, characterized in that: The determining whether to start image detection based on the real-time monitoring data and the radio frequency identification identification information includes: Comparing the real-time monitoring data with a minimum data threshold, and determining to start the image detection when the real-time monitoring data is less than the minimum data threshold and the radio frequency identification identification information is consistent with the train preset information; When the real-time monitoring data is greater than or equal to the minimum data threshold, or when the radio frequency identification identification information is inconsistent with the train preset information, it is determined that the image detection is not enabled.

3. The laser radar-based train track detection method according to claim 2, characterized in that: The image data after the switch switching is processed based on Canny edge detection to obtain the image data of the track connection, including: Performing Gaussian filtering on the image data after the switch switching, and adjusting the standard deviation of the Gaussian kernel to smooth the image; Calculating the gradient amplitude and direction of the image data after the switch switching, calculating the gradients in the horizontal and vertical directions based on the Sobel operator, synthesizing the total gradient G based on the gradients in the horizontal and vertical directions, and obtaining the gradient direction; Based on non-maximum suppression, adjacent pixels are checked according to the gradient direction, and only pixels with the local maximum gradient value are retained to remove redundant information; Based on double critical value detection, the edge strength is classified to determine the points to be retained and removed; Edge connection is performed according to the reserved points, and possible edges are tracked. If the possible edge is connected to a strong edge, it is retained; otherwise, it is removed, and finally the image data of the track connection is obtained.

4. The laser radar-based train track detection method according to claim 3, characterized in that: Based on dual-threshold detection, edge strength is classified to determine whether to retain or remove points, including: A high critical value Th and a low critical value Tl are set. When the total gradient G of the edge pixel is greater than the high critical value Th, the edge pixel is defined as a strong edge and retained. When Th≥G≥Tl, the edge pixel is defined as a possible edge and retained. When the total gradient G of the edge pixel is less than the low critical value Tl, the edge pixel is defined as a weak edge and removed.

5. The laser radar-based train track detection method according to claim 4, characterized in that: Determining the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and judging whether there is a switch switching abnormality, including: Marking all track positions according to the track connection image data, and arbitrarily selecting track positions to obtain several track gauges, and obtaining the straight line angle after the switch is switched according to the track connection image data; Comparing the straight line angle with a preset switch switching angle, and comparing all the track gauges with the standard track gauge, and determining whether there is a switch switching anomaly based on the comparison results; When the straight line angle is equal to the preset switch switching angle, and the track gauges of all the tracks are equal to the standard track gauge, it is determined that there is no switch switching abnormality; When the straight line angle is not equal to the preset switch switching angle, or the track gauge is not equal to the standard track gauge, it is determined that a switch switching abnormality exists.

6. The laser radar-based train track detection method according to claim 5, characterized in that: Determining the deviation value according to the straight line angle after the switch switching and the track gauge after the switch switching includes: Obtaining an angle deviation ratio based on the straight line angle after the switch switching and a preset switch switching angle. When obtaining the angle deviation ratio, calculating the absolute value of the difference between the straight line angle after the switch switching and the preset switch switching angle, obtaining a ratio of the absolute value of the difference to the preset switch switching angle, and using the angle ratio as the angle deviation ratio. Obtaining a track gauge deviation ratio based on the track gauge after the switch switching and the standard track gauge. When obtaining the track gauge deviation ratio, extracting the track gauge corresponding to the maximum difference between the track gauge and the standard track gauge, obtaining a ratio of the maximum difference to the standard track gauge, and using the track gauge ratio as the track gauge deviation ratio; Obtain a total deviation ratio based on the angle deviation ratio and the gauge deviation ratio, where the total deviation ratio is the sum of the angle deviation ratio and the gauge deviation ratio, compare the total deviation ratio with a first-level deviation ratio threshold and a second-level deviation ratio threshold, respectively, and determine a deviation value based on the comparison results, where the first-level deviation ratio threshold is less than the second-level deviation ratio threshold; When the total deviation ratio is less than or equal to the first-level deviation ratio threshold, the deviation value is determined to be a first-level deviation value; when the total deviation ratio is greater than the first-level deviation ratio threshold and less than or equal to the second-level deviation ratio threshold, the deviation value is determined to be a second-level deviation value; when the total deviation ratio is greater than the second-level deviation ratio threshold, the deviation value is determined to be a third-level deviation value; and the first-level deviation value is less than the second-level deviation value, and the second-level deviation value is less than the third-level deviation value.

7. The laser radar-based train track detection method according to claim 6, characterized in that: Determining whether to adjust the deviation value according to the image data of the track connection includes: Determine whether there is debris blocking the track joint according to the joint image data, and determine whether to adjust the deviation value according to the debris situation; When there are debris at the track joint, it is determined that the deviation value should be adjusted; When there is no debris at the track joint, it is determined that the deviation value is not adjusted, and the deviation value is used as the final deviation value.

8. The laser radar-based train track detection method according to claim 7, characterized in that: Determining whether to adjust the deviation value includes: Obtain debris size data, determine a correction coefficient based on the debris size data, adjust the deviation value to determine the final deviation value, the correction coefficient is proportional to the debris size data, and the final deviation value is the product of the deviation value and the correction coefficient, and the correction coefficient ranges from 1 to 1.5, where the correction coefficient includes 1.5 and excludes 1.

9. A laser radar-based train track detection system, used to apply the laser radar-based train track detection method according to any one of claims 1 to 8, characterized in that: include: A data acquisition module is configured to collect real-time monitoring data of the laser radar and determine whether to enable image detection based on the real-time monitoring data and radio frequency identification identification information; a data processing module configured to, when it is determined that the image detection is enabled, collect image data of the switch switching action, determine image data after the switch switching, process the image data after the switch switching based on Canny edge detection, and obtain image data of the track connection; a state determination module configured to determine the straight line angle after the switch switching and the track gauge after the switch switching based on the image data of the track connection, and to determine whether there is a switch switching abnormality; The deviation correction module is configured to determine the deviation value according to the straight line angle after the switch switching and the track gauge after the switch switching when the state judgment module determines that an abnormality exists, and determine whether to adjust the deviation value according to the image data of the track connection to determine the final deviation value.

10. The laser radar-based train track detection system according to claim 9, characterized in that: Also includes: The alarm module is configured to determine an alarm level according to the final deviation value, and the alarm level is proportional to the final deviation value.

Images