Steel rail fastener detection method and system, storage medium and electronic equipment

Abstract

The invention provides a steel rail fastener detection method and system, a storage medium and electronic equipment. The method comprises the steps that a three-dimensional point cloud picture at a track is acquired; taking the highest point of each position of the three-dimensional point cloud picture as a reference plane so as to calculate a corresponding earth surface plane; finding out a point cloud representing a steel rail to obtain a steel rail point cloud picture; the steel rail point cloud picture comprises two steel rail point cloud blocks which are oppositely arranged; finding out an interference point cloud block representing a non-steel rail body; the method comprises the following steps: scanning a track by adopting a track inspection instrument to obtain a chord data graph, and finding out a measurement point cloud block representing a fastener based on the chord data graph; calculating the geometric position relationship between each data point in the measurement point cloud block and the reference plane and the earth surface plane, and the geometric position relationship between the adjacent data points, and determining the type of the fastener; and regularly monitoring and measuring the geometric position relationship between each data point in the point cloud block relative to the reference plane and the earth surface plane and the geometric position relationship between the adjacent data points, and detecting the integrity of each fastener.

Description

Technical Field

[0001] The present invention relates to the technical field of track detection, and particularly to a method, system, storage medium and electronic device for detecting rail fasteners. Background Art

[0002] In recent years, the rail transit industry has developed rapidly. The operating speed of the rail transit system has increased, the construction mileage has increased, and the lines have become increasingly busy. This has led to an increase in the pressure on the track. Generally, the track is composed of multiple sections of rails connected in sequence, and the joints between two rails are connected by fasteners. Therefore, the state of the fasteners needs to be detected frequently to ensure the safety of the track.

[0003] In the prior art, for the defects of fasteners, an image processing-based visual solution is generally adopted, and a large amount of data is input to train a defect recognition model. This results in a large amount of model training, and the detection accuracy is greatly affected by ambient light and the quality of the model, thus causing various errors in the detection process. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method for detecting rail fasteners, aiming to solve the technical problems in the prior art that for the defects of fasteners, an image processing-based visual solution is generally adopted, and a large amount of data is input to train a defect recognition model. This results in a large amount of model training, and the detection accuracy is greatly affected by ambient light and the quality of the model, thus causing various errors in the detection process.

[0005] In order to achieve the above purpose, the present invention is implemented by the following technical solutions:

[0006] A method for detecting rail fasteners includes the following steps:

[0007] Use a lidar to scan the track in sequence to obtain a three-dimensional point cloud map of the track;

[0008] Regard the highest point at each position of the three-dimensional point cloud map as a reference plane to calculate the corresponding ground surface plane;

[0009] Perform feature analysis on the three-dimensional point cloud map to find the point cloud representing the rail to obtain a rail point cloud map; the rail point cloud map includes two relatively arranged rail point cloud blocks;

[0010] Analyze the vertical and horizontal positions of the rail point cloud blocks to find the interfering point cloud blocks representing non-the rail body;

[0011] Use an orbital inspection instrument to scan the track to obtain a chord data map, and based on the chord data map, find the interfering point cloud blocks representing the fasteners and use them as measurement point cloud blocks;

[0012] Calculate the geometric position relationship between each data point in the measured point cloud block with respect to the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, so as to determine the type of the fastener;

[0013] Regularly monitor the geometric position relationship between each data point in the measured point cloud block with respect to the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, to detect the integrity of each fastener.

[0014] According to one aspect of the above technical solution, after using the lidar to scan the track in sequence to obtain a three-dimensional point cloud map of the track, it further includes:

[0015] Denoise the three-dimensional point cloud map by means of direct filtering method and voxel filtering method to reduce the data points of the three-dimensional point cloud map.

[0016] According to one aspect of the above technical solution, regarding taking the highest point at each position of the three-dimensional point cloud map as the reference plane to calculate the corresponding ground surface plane, it specifically includes:

[0017] Arbitrarily select three points in sequence at the highest points of the three-dimensional point cloud map to construct multiple reference planes, extract all the data points on the reference plane as top surface data points, and take all the data points in the vertical direction of the top surface data points as pending data points;

[0018] Set a first height threshold, and calculate the first vertical height from each of the top surface data points to each of the pending data points in its vertical direction;

[0019] Calculate whether the first vertical height is greater than the first height threshold. If so, take the pending data points as ground data points;

[0020] Introduce an optimization formula, and calculate whether the number of the ground data points meets a predetermined number based on the current environment where the track is located;

[0021] If the prediction probability is less than the preset probability, re-select three points to construct the reference plane, and perform iterative calculation of the number of ground data points until the prediction probability is greater than the preset probability;

[0022]

[0023] Wherein, P represents the prediction probability, γ represents the predetermined number of ground data points, θ represents the number of ground data points that exceed or are lacking, q represents the number of all pending data points in this vertical direction, and Q represents the number of iterations;

[0024] Take the plane composed of the highest ground data points obtained from the last calculation as the ground surface plane.

[0025] According to one aspect of the above technical solution, perform feature analysis on the three-dimensional point cloud map to find the point cloud representing the rail to obtain a rail point cloud map; the rail point cloud map includes two relatively arranged rail point cloud blocks, specifically including:

[0026] Taking the ground surface plane as the reference plane and the extending direction of the rail in the track as the X-axis, a plurality of starting points x in the X-axis are proposed n , and a sliding window (x n , 0) is set;

[0027] According to the preset width of the rail as the search command, slide the sliding window along the width direction Y-axis of the rail to obtain a plurality of rail bottom pane points;

[0028] (w 11 , w 22 , w 33 ......w nm )

[0029] wherein, w nm represents the m-th rail bottom pane point starting from the n-th starting point x n ;

[0030] Extract all the data points above the rail bottom pane points, regarded as rail body data points, to obtain a rail point cloud map, and then extract two rail point cloud blocks from the rail point cloud map.

[0031] According to one aspect of the above technical solution, analyze the vertical position and horizontal position of the rail point cloud block to find the interference point cloud block representing non-the rail body, specifically including:

[0032] Calculate the second vertical height of each data point of the rail point cloud block with its left adjacent point and right adjacent point;

[0033] H ij1 =|z ij -z i(j+1) |

[0034] H ij2 =|z ij -z i(j-1) |;

[0035] wherein, Z ij is the height of the data point at the i-th position of the X-axis and the j-th position of the Y-axis, H ij1 is the second vertical height of the data point with its right adjacent point, and H ij2 is the second vertical height of the data point with its left adjacent point;

[0036] Set a second height threshold. If the second vertical height is between the second height thresholds, consider the left adjacent point or the right adjacent point as a pending ballast point cloud block;

[0037] Import the historical data at the pending ballast point cloud block and evaluate the credibility of the pending ballast point cloud block;

[0038]

[0039] where t is the time, k is the number of measurements, Z tij is the height of the data point at the ij position at time t, is the average height of the data points at the ij position within the measurement time range, and δ ij is the standard deviation of the height of the data points at the ij position;

[0040] If the data points in the pending ballast point cloud block meet the credibility determination criteria, consider the pending ballast point cloud block as a ballast point cloud block and exclude the ballast point cloud block from the rail point cloud block;

[0041]

[0042] where μ is the error range, and 2 ≤ μ ≤ 3;

[0043] In the updated rail point cloud block, calculate the horizontal distance between the edge data points at each horizontal height. Refer to the rail design standard, and put forward the edge data points with the horizontal distance greater than the rail design standard and consider them as pending foreign object point cloud blocks;

[0044] Import the historical data at the pending foreign object point cloud block and evaluate the credibility of the pending foreign object point cloud block;

[0045]

[0046] where X tij is the longitudinal horizontal position of the data point at the ij position at time t, is the average longitudinal horizontal position of the data points at the ij position within the measurement time range, and Δ ij is the standard deviation of the longitudinal horizontal position of the data points at the ij position;

[0047] If the data points in the pending foreign object point cloud block meet the credibility determination criteria, consider the pending foreign object point cloud block as a foreign object point cloud block;

[0048]

[0049] Consider the ballast point cloud block and the foreign object point cloud block together as interference point cloud blocks.

[0050] According to one aspect of the above technical solution, the track is scanned by a track inspection instrument to obtain a chord data map. Based on the chord data map, interference point cloud blocks representing fasteners are found and used as measurement point cloud blocks. Specifically, it includes:

[0051] The track is scanned by a track inspection instrument to obtain a chord data map, and a chord data matrix T is constructed according to the chord data map;

[0052]

[0053] where N is the total number of scanning samples by the track inspection instrument, is the scanning sample data;

[0054] Based on the matrix binary recurrence construction method, the chord data matrix is subjected to singular value decomposition to obtain similar singular values and key singular values respectively, and then the similar signals ε and key signals λ corresponding to the similar singular values and the key singular values are obtained respectively;

[0055] The similar signal is decomposed step by step to obtain the similar signal ε at each level a and the key signal λ a ; where a is the number of decomposition times;

[0056] A key signal matrix D is constructed according to all levels of the key signals a ;

[0057] In the key signal matrix D a find the signals with peak mutations, and regard the signals with peak mutations as the first pending signals;

[0058] Analyze the signals with the same peaks and consistent spacings in the first pending signals, and regard them as the second pending signals;

[0059] Referring to the fastener design specifications, regard the second pending signals with spacings conforming to the fastener design specifications as fastener signals, and regard the interference point cloud blocks at the positions of the fastener signals as measurement point cloud blocks.

[0060] According to one aspect of the above technical solution, calculate the geometric position relationship between each data point in the measurement point cloud block and the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, so as to determine the type of the fastener. Specifically, it includes:

[0061] Calculate the vertical distances between each data point in the measurement point cloud block and the reference plane and the ground surface plane respectively. Referring to the fastener design specifications, regard the data points with vertical distances conforming to the fastener design specifications as the data points to be inspected;

[0062] Hij3 =z ij

[0063] H ij4 =|z ij -z′ ij |;

[0064] Among them, z′ ij is the height of the reference plane at the i-th position of the X axis and the j-th position of the Y axis, H ij3 is the vertical distance between the data point and the ground plane, H ij4 is the vertical distance between the data point and the reference plane;

[0065] Calculate the slope between adjacent data points to be tested, and refer to the fastener design specification to find the corresponding fastener type;

[0066]

[0067] Among them, Yij is the horizontal position of the data point at position ij, G ij1 is the slope between the data point and the previous adjacent point, G ij2 is the slope between a data point and its adjacent point.

[0068] The present invention also provides a rail fastener detection system, comprising:

[0069] The first scanning module is used to scan the track in sequence using a laser radar to obtain a three-dimensional point cloud map of the track;

[0070] Denoising module: used for performing denoising processing on the three-dimensional point cloud image by a straight-through filtering method and a voxel filtering method to reduce the data points of the three-dimensional point cloud image;

[0071] Calculation module: used for considering the highest point of each position of the three-dimensional point cloud image as a reference plane to calculate the corresponding ground surface plane;

[0072] The first analysis module is used to perform feature analysis on the three-dimensional point cloud image to find out the point cloud representing the rails to obtain a rail point cloud image; the rail point cloud image includes two rail point cloud blocks arranged relatively;

[0073] The second analysis module is used to analyze the vertical position and horizontal position of the rail point cloud block to find out the interference point cloud block that does not represent the rail body;

[0074] The second scanning module is used to scan the track using a track inspection instrument to obtain a chord data map, and based on the chord data map, find out the interference point cloud block representing the fastener and use it as the measurement point cloud block;

[0075] Determination module: used to calculate the geometric positional relationships between each data point in the measured point cloud block with respect to the reference plane and the ground surface plane, as well as the geometric positional relationships between adjacent data points, so as to determine the type of the fastener;

[0076] Monitoring module: used to regularly monitor the geometric positional relationships between each data point in the measured point cloud block with respect to the reference plane and the ground surface plane, as well as the geometric positional relationships between adjacent data points, so as to detect the integrity of each fastener.

[0077] The present invention also provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the rail fastener detection method as described above is implemented.

[0078] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the rail fastener detection method as described above is implemented.

[0079] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0080] By scanning the track with a lidar, a three-dimensional point cloud map of the track can be obtained. Then, the reference plane and the ground surface plane of the three-dimensional point cloud map are found in the vertical layer. At this time, the three-dimensional point cloud map also includes ballast, weeds, etc. near the rail. Therefore, by performing feature analysis on the three-dimensional point cloud map, first exclude the data points representing ballast and weeds to obtain the rail point cloud block. At this time, the rail point cloud block still has data points representing fasteners and foreign objects on the rail. Therefore, by analyzing the vertical and horizontal positions of the rail point cloud block and referring to the dimensions during rail design, fasteners, foreign objects, etc. that do not meet the design dimensions can be regarded as interfering point cloud blocks; since the amount of data points of fasteners and foreign objects is large, a chord data map is obtained by scanning the track with a track inspection instrument to exclude other foreign objects except fasteners to obtain a measured point cloud block; analyze the positional relationships between the data points in the measured point cloud block and the reference plane and the ground surface plane, as well as the positional relationships between the data points in the measured point cloud block, and refer to the dimensions during fastener design to determine the type of the fastener; subsequently, only need to analyze the missing situation of each data point in the measured point cloud block to know the damage situation of the fastener, and then accurately and quickly detect the fastener.

[0081] The present invention determines the damage position of the fastener by means of point cloud. Compared with visual solutions such as image processing, this solution can greatly reduce the interference of the environment on the recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a flowchart of the rail fastener detection method in the first embodiment of the present invention;

[0083] Figure 2 It is a schematic structural diagram of the U-shaped fastener in the first embodiment of the present invention;

[0084] Figure 3 It is a structural block diagram of the rail fastener detection system in the second embodiment of the present invention;

[0085] Figure 4 It is a structural block diagram of the electronic device in the third embodiment of the present invention;

[0086] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. Specific Embodiments

[0087] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0088] It should be noted that when an element is referred to as being "fixedly provided on" another element, it can be directly on the other element or there may also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0089] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0090] Please refer to Figure 1 , which shows a rail fastener detection method in the first embodiment of the present invention, including the following steps:

[0091] S10. Use a lidar to scan the track in sequence to obtain a three-dimensional point cloud map of the track;

[0092] S20. Consider the highest point at each position of the three-dimensional point cloud map as the reference plane to calculate the corresponding ground surface plane;

[0093] S30. Perform feature analysis on the three-dimensional point cloud map to find the point cloud representing the rail to obtain a rail point cloud map; the rail point cloud map includes two relatively arranged rail point cloud blocks;

[0094] S40, analyzing the vertical position and the horizontal position of the rail point cloud block to find out the interference point cloud block representing the non-rail body;

[0095] S50, scanning the track using a track inspection instrument to obtain a chord data graph, and based on the chord data graph, finding an interference point cloud block representing a fastener, and using the interference point cloud block as a measurement point cloud block;

[0096] S60, calculating the geometric position relationship between each data point in the measured point cloud block and the reference plane and the ground surface plane, and the geometric position relationship between adjacent data points, so as to determine the type of the fastener;

[0097] S70, regularly monitoring the geometric position relationship between each data point in the measurement point cloud block and the reference plane and the ground surface plane to detect the integrity of each fastener.

[0098] It can be understood that the present invention can obtain a three-dimensional point cloud map at the track by scanning the track with a laser radar, and then find the reference plane and the surface plane of the three-dimensional point cloud map in the vertical plane. At this time, the three-dimensional point cloud map also includes ballast, weeds, etc. near the rails. Therefore, the three-dimensional point cloud map is subjected to feature analysis, and the data points represented by the ballast and weeds are first excluded to obtain a rail point cloud block. At this time, the rail point cloud block also has data points represented by fasteners and foreign objects on the rails. Therefore, by analyzing the vertical and horizontal positions of the rail point cloud block, referring to the size of the rail when it was designed, the points that do not meet the design size can be removed. Fasteners and foreign objects of different sizes are considered as interference point cloud blocks; since the number of data points of fasteners and foreign objects is large, the track is scanned by a track inspection instrument to obtain a chord data map to exclude other foreign objects except fasteners to obtain a measurement point cloud block; the positional relationship between the data points in the measurement point cloud block and the reference plane and the surface plane, as well as the positional relationship between the data points in the measurement point cloud block, are analyzed, and the type of fastener is determined by referring to the dimensions of the fastener when it was designed; subsequently, it is only necessary to analyze the missing status of each data point in the measurement point cloud block to know the damage status of the fastener, and then accurately and quickly detect the fastener.

[0099] The present invention determines the damaged position of the fastener by means of point cloud. Compared with visual solutions such as image processing, this solution can greatly reduce the interference of the environment on the recognition accuracy.

[0100] Specifically, in this embodiment, after step S10, the following steps are further performed:

[0101] The three-dimensional point cloud image is denoised by a straight-through filtering method and a voxel filtering method to reduce the data points of the three-dimensional point cloud image.

[0102] It can be understood that the direct filtering method sets a range on three axes to preliminarily filter the data points in the specified range in the three-dimensional point cloud map. For example, there is generally a preliminary range for the position of weeds. We set the preliminary range representing weeds as the range to be filtered by the direct filtering method for rough filtering. The voxel filtering method, on the other hand, performs downsampling on the three-dimensional point cloud map to reduce the number of data points and increase the processing speed for data points.

[0103] Further, the specific steps of step S20 include:

[0104] Arbitrarily select three points in sequence at the highest point of the three-dimensional point cloud map to construct multiple reference planes, extract each data point on the reference plane as the top surface data point, and use all data points in the vertical direction of the top surface data point as the pending data points;

[0105] Set a first height threshold, and calculate the first vertical height from each of the top surface data points to each of the pending data points in its vertical direction;

[0106] Calculate whether the first vertical height is greater than the first height threshold. If so, use the pending data point as a ground data point;

[0107] Introduce an optimization formula, and calculate whether the number of the ground data points meets a predetermined number based on the environment where the track is currently located;

[0108] If the predicted probability is less than the preset probability, re-select three points to construct the reference plane and perform iterative calculation of the number of ground data points until the predicted probability is greater than the preset probability;

[0109]

[0110] Where P represents the predicted probability, γ represents the predetermined number of ground data points, θ represents the number of ground data points that exceed or are lacking, q represents the number of all pending data points in this vertical direction, and Q represents the number of iterations;

[0111] Use the plane composed of the highest ground data points obtained from the last calculation as the ground surface plane.

[0112] It can be understood that in the three-dimensional point cloud map, the highest point on the Z-axis is generally the upper surface of the rail. When the rail is designed, its height is fixed. Therefore, along the extension direction of the rail, three points are arbitrarily selected in sequence at the highest point of the three-dimensional point cloud map (that is, the points on the upper surface of the rail) to construct a reference plane. There are multiple reference planes, and each reference plane is represented by a plane equation. The points on the reference plane are called top surface data points. Then, the first vertical height from the top surface data points to all the to-be-determined data points below is calculated. If the first vertical height is greater than the first height threshold (the first height threshold is the vertical height of the rail), it means that the to-be-determined data point is a ground data point. To avoid the influence of the environment on the number of data points, for example, if it is just at the weld here, or there is a foreign object on the rail here, which makes one of the points of the reference plane constructed by the three selected points relatively high, resulting in a large number of finally calculated ground data points, or if the rail has a defect, which makes one of the points of the reference plane constructed by the three selected points relatively low, resulting in a small number of finally calculated ground data points. Therefore, it is necessary to re-determine the reference plane to calculate whether the number of ground data points meets the predetermined number until the prediction probability in the optimization formula is greater than the preset probability, and the plane composed of the highest ground data points obtained in the last calculation is used as the ground surface plane. In this way, the upper and lower planes of the rail are determined.

[0113] Further, the specific steps of step S30 include:

[0114] Taking the ground surface plane as the reference plane and the extension direction of the rail in the track as the X-axis, multiple starting points x in the X-axis are proposed n , and a sliding window (x n , 0) is set;

[0115] According to the preset width of the rail as the search command, the sliding window is slid along the width direction Y-axis of the rail to obtain the rail bottom pane points;

[0116] (w 11 , w 22 , w 33 ......w nm )

[0117] where w nm represents the mth rail bottom pane point starting from the nth starting point x n ;

[0118] All the data points above the rail bottom pane points are extracted and regarded as rail body data points to obtain the rail point cloud map, and then two rail point cloud blocks are extracted from the rail point cloud map.

[0119] It can be understood that after calculating the reference plane and the ground surface plane, it is necessary to obtain the position of the actual point cloud block of the rail in the three-dimensional point cloud map. By taking the ground surface plane as the reference plane, setting multiple starting points along the extension direction of the rail, and taking the vertical direction of the rail as the sliding direction and the width of the rail as the search command, the window points on the bottom surface of the rail are searched. Since the sizes of each cross-section of the rail are inconsistent, all the window points on the window points on the bottom surface of the rail are regarded as the window points of the rail body, and then the rail point cloud map and the point cloud blocks of the two rails are obtained.

[0120] Further, the specific steps of step S40 include:

[0121] Calculate the second vertical height of each data point of the rail point cloud block with its left adjacent point and right adjacent point;

[0122] H ij1 =|z ij -z i(j+1) |

[0123] H ij2 =|z ij -z i(j-1) |;

[0124] Wherein, Z ij is the height of the data point at the i-th position of the X-axis and the j-th position of the Y-axis, H ij1 is the second vertical height of the data point with its right adjacent point, and H ij2 is the second vertical height of the data point with its left adjacent point;

[0125] Set the second height threshold. If the second vertical height is between the second height thresholds, the left adjacent point or the right adjacent point is regarded as a pending ballast point cloud block;

[0126] Import the historical data at the pending ballast point cloud block, and evaluate the credibility of the pending ballast point cloud block;

[0127]

[0128] Wherein, t is the time, k is the number of measurements, Z tij is the height of the data point at the ij position at time t, is the average height of the data points at the ij position within the measurement time range, and δ ij is the height standard deviation of the data point at the ij position;

[0129] If the data points in the pending ballast point cloud block meet the criteria for credibility determination, the pending ballast point cloud block is regarded as a ballast point cloud block, and the ballast point cloud block is excluded from the rail point cloud block;

[0130]

[0131] Among them, μ is the error range, where 2 ≤ μ ≤ 3;

[0132] In the updated rail point cloud block, calculate the horizontal distance between the edge data points at each horizontal height. Referring to the rail design standard, extract the edge data points whose horizontal distance is greater than the rail design standard and regard them as the pending foreign object point cloud blocks;

[0133] Import the historical data at the location of the pending foreign object point cloud block and conduct a credibility assessment on the pending foreign object point cloud block;

[0134]

[0135] Among them, X tij is the longitudinal horizontal position of the data point at the ij position at time t, is the average longitudinal horizontal position of the data points at the ij position within the measurement time range, and Δ ij is the standard deviation of the longitudinal horizontal position of the data points at the ij position;

[0136] If the data points in the pending foreign object point cloud block meet the credibility determination criteria, then regard the pending foreign object point cloud block as a foreign object point cloud block;

[0137]

[0138] Regard the ballast point cloud block and the foreign object point cloud block together as interference point cloud blocks.

[0139] It can be understood that the very bottom end of the rail is very close to the ground but has a certain distance, and there is ballast beside the rail. The ballast is very close to the rail and even covers the web of the rail. In order to eliminate the interference of the ballast on the rail point cloud block, calculate the second vertical height of each data point in the rail point cloud block with its left adjacent point and right adjacent point, and calculate whether the second vertical height with the left adjacent point and right adjacent point is between the second height thresholds (generally the distance from the lower flange to the road surface and the distance from the upper flange to the lower flange). Generally, the data points between each data point in the rail point cloud block and its left adjacent point and right adjacent point are continuous. If there is a sudden change in height, either a foreign object or ballast is encountered, or the upper flange has dropped to the lower flange, or the lower flange has dropped to the road surface. If it is not between this second height threshold but there is a sudden change in height, it means a foreign object or ballast has been encountered, and these data points should be eliminated;

[0140] Since it is common for ballast to spill onto the rails, the workload of removing these data points is too large. Therefore, historical data is introduced to check the values of the second vertical height at this location within a historical time period. If it does not meet the credibility judgment criteria, it means that the ballast at this location has fallen off the rail due to environmental reasons, or the ballast at this location has become ballast of another height (indicating that the rail at this location is prone to accumulating ballast). There is no need to remove it, and it will fall off over time. If it meets the credibility judgment criteria, it means that the ballast at this location is very firm and needs to be removed, excluding the ballast point cloud block;

[0141] Since the distances between the cross-sections at various heights of the rail are determined by the rail design standard, if the horizontal distance of a cross-section is greater than the rail design standard, it means there is a foreign object here. According to the above steps for evaluating the credibility of ballast, historical data is introduced to evaluate the credibility of the foreign object. If it does not meet the credibility judgment criteria, it means the foreign object is prone to falling off and can be ignored. If it meets the credibility judgment criteria, it should be regarded as a foreign object point cloud block and removed. Then, the ballast point cloud block and the foreign object point cloud block are jointly regarded as interference point cloud blocks.

[0142] Furthermore, the specific steps of step S50 include:

[0143] Use a track inspection instrument to scan the track to obtain a chord data diagram, and construct a chord data matrix T based on the chord data diagram;

[0144]

[0145] where N is the total number of scanning samples by the track inspection instrument, is the scanning sample data;

[0146] Based on the matrix binary recurrence construction method, perform singular value decomposition on the chord data matrix to obtain similar singular values and key singular values respectively, and then obtain the similar signal ε and key signal λ corresponding to the similar singular value and the key singular value respectively;

[0147] Perform hierarchical decomposition on the similar signal to obtain the similar signal ε at each level a and the key signal λ a ; where a is the number of decomposition times;

[0148] Construct a key signal matrix D based on all levels of the key signals a ;

[0149] In the key signal matrix D a find the signal with a peak mutation and regard the signal with a peak mutation as the first signal to be determined;

[0150] Analyze the signals with the same peaks and consistent intervals in the first signal to be determined, and regard them as the second signal to be determined;

[0151] Refer to the fastener design specification, regard the second signal to be determined with an interval conforming to the fastener design specification as the fastener signal, and regard the interference point cloud block at the position of the fastener signal as the measurement point cloud block.

[0152] It can be understood that since it is difficult to accurately distinguish between foreign objects and fasteners from the point cloud map, an orbit inspection instrument is introduced to scan the orbit to obtain a chord data map, and then a chord data matrix H is constructed. The chord data matrix H is subjected to singular value decomposition based on the matrix binary recursive construction method. When encountering foreign objects, the chord signal will have peaks (i.e., singular values), and rail unevenness will also have peaks. In order to distinguish whether the peak is caused by unevenness, foreign objects, or fasteners, the chord data matrix H is subjected to singular value decomposition, and similar singular values and key singular values can be obtained, and then similar signals and key signals can be obtained. Similar signals are generally caused by unevenness and are not intense, so they look similar. Key signals reflect the peaks caused by foreign objects, fasteners, etc. Through hierarchical decomposition, a more pure key signal can be obtained, and then a key signal matrix D can be obtained. j , signals with peak mutations can be found from the key signal matrix D j The peaks reflected by foreign objects have a characteristic that the peaks between adjacent foreign objects are inconsistent and the intervals are not equal, so such signals can be excluded. The peaks caused by welds and fasteners both have a characteristic that the peaks are consistent and the intervals are equal. Therefore, according to the interval specification between fasteners in the fastener design specification, the signal representing the fastener is found, and the signal representing the weld is excluded, and then the interference point cloud block at the same position is regarded as the measurement point cloud block.

[0153] Further, the specific steps of step S60 include:

[0154] Calculate the vertical distances between each data point in the measurement point cloud block and the reference plane and the ground surface plane respectively. Refer to the fastener design specification, and regard the data points with vertical distances conforming to the fastener design specification as the data points to be inspected;

[0155] H ij3 = z ij

[0156] H ij4 = |z ij - z′ ij |;

[0157] where z′ ij is the height of the reference plane of the X-axis at the i-th position and the Y-axis at the j-th position, Hij3 is the vertical distance between the data point and the ground plane, H ij4 is the vertical distance between the data point and the reference plane;

[0158] Calculate the slope between adjacent data points to be tested, and refer to the fastener design specification to find the corresponding fastener type;

[0159]

[0160] Among them, Yij is the horizontal position of the data point at position ij, G ij1 is the slope between the data point and the previous adjacent point, G ij2 is the slope between a data point and its adjacent point.

[0161] It can be understood that, taking the U-shaped fastener as an example, its shape is as follows Figure 2 As shown, the fastener spring bar feature is U-shaped, and the fastener bolt does not contact the spring bar. We calculate the vertical distance between each data point and the reference plane and the ground plane in the measurement point cloud block representing the U-shaped fastener. The vertical distance between the lower side of the U-shape and the ground plane, the vertical distance between the upper side of the U-shape and the reference plane, and the vertical distance between the right side of the U-shape and the reference plane and the ground plane can all be calculated. According to the vertical distance between each data point and the reference plane and the ground plane, we can preliminarily consider that the measurement point cloud block represents a U-shaped fastener and regard it as a data point to be tested. Then we focus on calculating the slope between the right side of the U-shape and the lower side of the U-shape and the upper side of the U-shape, especially at the corners. We can basically determine that the fastener type is a U-shaped fastener.

[0162] Afterwards, by repeating steps S10 to S60 irregularly, the measurement point cloud block can be updated, and the integrity of each fastener can be detected based on the loss or addition (foreign matter on the fastener) of each data point in the measurement point cloud block, thereby comprehensively monitoring the fasteners.

[0163] In summary, the rail fastener detection method in the above-mentioned embodiment of the present invention determines the damaged position of the fastener by means of point cloud. Compared with visual solutions such as image processing, this solution can greatly reduce the interference of the environment on the recognition accuracy.

[0164] Please refer to Figure 3 , shown is a rail fastener detection system in a second embodiment of the present invention, comprising:

[0165] The first scanning module 11 is used to sequentially scan the track using a laser radar to obtain a three-dimensional point cloud image of the track;

[0166] De-noising module 12: used for performing denoising processing on the three-dimensional point cloud image by a straight-through filtering method and a voxel filtering method to reduce the data points of the three-dimensional point cloud image;

[0167] Calculation module 13: used to regard the highest point at each position of the three-dimensional point cloud map as a reference plane to calculate the corresponding ground surface plane;

[0168] Specifically, the calculation module 13 is used for:

[0169] Arbitrarily select three points in sequence at the highest points of the three-dimensional point cloud map to construct multiple reference planes, extract each data point on the reference plane as a top surface data point, and regard all data points in the vertical direction of the top surface data point as undetermined data points;

[0170] Set a first height threshold, and calculate the first vertical height from each of the top surface data points to each of the undetermined data points in its vertical direction;

[0171] Calculate whether the first vertical height is greater than the first height threshold. If so, regard the undetermined data point as a ground data point;

[0172] Introduce an optimization formula, and calculate whether the number of the ground data points meets a predetermined number based on the environment where the track is currently located;

[0173] If the prediction probability is less than the preset probability, re-select three points to construct the reference plane, and perform iterative calculation on the number of ground data points until the prediction probability is greater than the preset probability;

[0174]

[0175] Wherein, P represents the prediction probability, γ represents the predetermined number of ground data points, θ represents the number of ground data points exceeding or lacking, q represents the number of all undetermined data points in this vertical direction, and Q represents the number of iterations;

[0176] Regard the plane composed of the highest ground data points obtained from the last calculation as the ground surface plane;

[0177] First analysis module 14: used to perform feature analysis on the three-dimensional point cloud map, find out the point cloud representing the rail to obtain a rail point cloud map; the rail point cloud map includes two relatively arranged rail point cloud blocks;

[0178] Specifically, the first analysis module 14 is used for:

[0179] Take the ground surface plane as the reference plane, take the extending direction of the rail in the track as the X axis, and extract multiple starting points x in the X axis n , and set a sliding window (x n , 0);

[0180] Using the preset width of the rail as a search command, slide the sliding window along the width direction (Y-axis) of the rail to obtain multiple rail bottom pane points;

[0181] (w 11 , w 22 , w 33 ......w nm )

[0182] where w nm represents the m-th rail bottom pane point starting from the n-th starting point x n ;

[0183] Extract all the data points above the rail bottom pane points, regarded as rail body data points, to obtain a rail point cloud map, and then extract two rail point cloud blocks from the rail point cloud map;

[0184] The second analysis module 15: used to analyze the vertical position and horizontal position of the rail point cloud blocks to find out the interfering point cloud blocks that do not represent the rail body;

[0185] Specifically, the second analysis module 15 is used for:

[0186] Calculate the second vertical height of each data point in the rail point cloud block with its left adjacent point and right adjacent point;

[0187] H ij1 =|z ij -z i(j+1) |

[0188] H ij2 =|z ij -z i(j-1) |;

[0189] where Z ij is the height of the data point at the i-th position of the X-axis and the j-th position of the Y-axis, H ij1 is the second vertical height of the data point with its right adjacent point, and H ij2 is the second vertical height of the data point with its left adjacent point;

[0190] Set a second height threshold. If the second vertical height is within the second height threshold, regard the left adjacent point or right adjacent point as a pending ballast point cloud block;

[0191] Import the historical data at the pending ballast point cloud block to evaluate the credibility of the pending ballast point cloud block;

[0192]

[0193] where t is the time, k is the number of measurements, Ztij The height of the data point at the ij position at time t The average height of the data points at the ij position within the measurement time range, δ ij The standard deviation of the height of the data points at the ij position

[0194] If the data points in the to-be-determined ballast point cloud block meet the criteria of credibility determination, then regard the to-be-determined ballast point cloud block as a ballast point cloud block, and exclude the ballast point cloud block from the rail point cloud block

[0195]

[0196] Where μ is the error range, 2 ≤ μ ≤ 3

[0197] In the updated rail point cloud block, calculate the horizontal distance between the edge data points at each horizontal height. Referring to the rail design standard, put forward the edge data points whose horizontal distance is greater than the rail design standard, and regard them as to-be-determined foreign object point cloud blocks

[0198] Import the historical data at the position of the to-be-determined foreign object point cloud block, and conduct a credibility assessment on the to-be-determined foreign object point cloud block

[0199]

[0200] Where X tij The longitudinal horizontal position of the data point at the ij position at time t The average longitudinal horizontal position of the data points at the ij position within the measurement time range, Δ ij The standard deviation of the longitudinal horizontal position of the data points at the ij position

[0201] If the data points in the to-be-determined foreign object point cloud block meet the criteria of credibility determination, then regard the to-be-determined foreign object point cloud block as a foreign object point cloud block

[0202]

[0203] Regard the ballast point cloud block and the foreign object point cloud block together as interference point cloud blocks

[0204] The second scanning module 16: used to scan the track by using an orbit inspection instrument to obtain a chord data diagram, and based on the chord data diagram, find out the interference point cloud blocks representing fasteners and take them as measurement point cloud blocks

[0205] The second scanning module 16 is specifically used for

[0206] Scan the track by using an orbit inspection instrument to obtain a chord data diagram, and construct a chord data matrix T according to the chord data diagram

[0207]

[0208] Among them, N is the total number of scanning samplings of the track inspection instrument, is the scanning sampling data;

[0209] Based on the matrix binary recurrence construction method, the chord data matrix is subjected to singular value decomposition to respectively obtain similar singular values and key singular values, and then the similar signals ε and key signals λ corresponding to the similar singular values and the key singular values are respectively obtained;

[0210] Perform hierarchical decomposition on the similar signals to obtain the similar signals ε at each level a and the key signal λ a ; where a is the number of decomposition times;

[0211] Construct a key signal matrix D according to the key signals at all levels a ;

[0212] In the key signal matrix D a Search for signals with peak mutations and regard the signals with peak mutations as the first pending signals;

[0213] Analyze the signals with the same peaks and consistent spacing in the first pending signals and regard them as the second pending signals;

[0214] Referring to the fastener design specification, regard the second pending signals with spacing conforming to the fastener design specification as fastener signals, and use the interference point cloud blocks at the positions of the fastener signals as the measurement point cloud blocks;

[0215] Determination module 17: Calculate the geometric position relationship between each data point in the measurement point cloud block and the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, so as to determine the type of the fastener;

[0216] The determination module 17 is specifically used for:

[0217] Calculate the vertical distances between each data point in the measurement point cloud block and the reference plane and the ground surface plane respectively, and refer to the fastener design specification to regard the data points with vertical distances conforming to the fastener design specification as the data points to be inspected;

[0218] H ij3 = z ij

[0219] H ij4 = |z ij - z' ij |;

[0220] Among them, z'ij is the height of the reference plane where the X-axis is at the i-th position and the Y-axis is at the j-th position, H ij3 is the vertical distance between the data point and the ground surface plane, H ij4 is the vertical distance between the data point and the reference plane;

[0221] Calculate the slope between adjacent to-be-tested data points, and refer to the fastener design specifications to find the corresponding fastener type;

[0222]

[0223] where Yij is the lateral horizontal position of the data point at the ij position, G ij1 is the slope between the data point and the previous adjacent point, G ij2 is the slope between the data point and the subsequent adjacent point;

[0224] Monitoring module 18: used to regularly monitor the geometric position relationship between each data point in the measured point cloud block with respect to the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, to detect the integrity of each fastener.

[0225] The present invention also proposes an electronic device, please refer to Figure 4 , which shows the electronic device in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored on the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned rail fastener detection method is implemented.

[0226] Among them, the memory 10 includes at least one type of storage medium, and the storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 10 can be an internal storage unit of the electronic device in some embodiments, such as the hard disk of the electronic device. The memory 10 can also be an external storage device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 10 can also include both the internal storage unit of the electronic device and the external storage device. The memory 10 can be used not only to store application software and various types of data installed in the electronic device, but also to temporarily store data that has been output or will be output.

[0227] Among them, in some embodiments, the processor 20 may be an Electronic Control Unit (ECU, also known as a vehicle computer), a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips, and is used to run the program code stored in the memory 10 or process data, such as executing an access restriction program and the like.

[0228] It should be noted that Figure 4 the structure shown does not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than shown in the figure, or combine certain components, or have different component arrangements.

[0229] An embodiment of the present invention also provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the rail fastener detection method as described above is implemented.

[0230] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0231] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection part (electronic device) having one or more wirings, a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0232] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0233] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0234] The above-described embodiments merely represent several implementation manners of the present invention, and the description thereof is relatively specific and detailed, but should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims

1. A rail fastener detection method, characterized in that: The steps include: The laser radar is used to scan the track in sequence to obtain a three-dimensional point cloud map of the track; The highest point at each position of the three-dimensional point cloud image is regarded as a reference plane to calculate the corresponding ground surface plane; Performing feature analysis on the three-dimensional point cloud image to find out the point cloud representing the rails, so as to obtain a rail point cloud image; the rail point cloud image includes two rail point cloud blocks arranged relatively; Analyze the vertical position and horizontal position of the rail point cloud block to find out the interfering point cloud block that does not represent the rail body; Scanning the track with a track inspection instrument to obtain a chord data graph, and finding an interference point cloud block representing a fastener based on the chord data graph, and using the interference point cloud block as a measurement point cloud block; Calculating the geometric position relationship between each data point in the measured point cloud block and the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, so as to determine the type of the fastener; The geometric position relationship between each data point in the measurement point cloud block and the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, are regularly monitored to detect the integrity of each fastener.

2. The rail fastener detection method according to claim 1, characterized in that: The laser radar is used to scan the track in sequence to obtain a three-dimensional point cloud map at the track, and then the following steps are further included: The three-dimensional point cloud image is denoised by a straight-through filtering method and a voxel filtering method to reduce the data points of the three-dimensional point cloud image.

3. The rail fastener detection method according to claim 2, characterized in that: The method of considering the highest point at each position of the three-dimensional point cloud image as a reference plane to calculate the corresponding ground surface plane specifically includes: Select three points at random at the highest point of the three-dimensional point cloud image in sequence to construct multiple reference planes, extract each data point on the reference plane as a top surface data point, and take all data points in a vertical direction of the top surface data point as pending data points; Setting a first height threshold, and calculating a first vertical height from each of the top surface data points to each of the pending data points in the vertical direction thereof; Calculate whether the first vertical height is greater than the first height threshold, and if so, use the pending data point as a ground data point; An optimization formula is introduced to calculate whether the number of the ground data points meets a predetermined number based on the current environment of the track; If the predicted probability is less than the preset probability, three points are reselected to construct the reference plane, and the number of ground data points is iteratively calculated until the predicted probability is greater than the preset probability; Where P represents the prediction probability, γ represents the number of predetermined ground data points, θ represents the number of exceeded or missing ground data points, q represents the number of all pending data points in the vertical direction, and Q represents the number of iterations; The plane formed by the highest ground data points obtained by the last calculation is taken as the ground surface plane.

4. The rail fastener detection method according to claim 3, characterized in that: The three-dimensional point cloud image is subjected to feature analysis to find out the point cloud representing the rails, so as to obtain a rail point cloud image; the rail point cloud image includes two rail point cloud blocks arranged relatively, specifically including: Taking the ground plane as the reference plane and the extension direction of the rail in the track as the X-axis, a plurality of starting points x in the X-axis are proposed. n , and set the sliding window (x n , 0); According to the preset width of the rail as a search command, the sliding window is slid along the Y-axis in the width direction of the rail to obtain a plurality of window points on the bottom surface of the rail; (In 11 ,In 22 In 33 ......In nm ) Among them, w nm Indicates that from the nth starting point x n The starting mth rail bottom pane point; All data points above the window point on the bottom surface of the rail are extracted and regarded as data points of the rail body to obtain a rail point cloud map, and then two rail point cloud blocks are extracted from the rail point cloud map.

5. The rail fastener detection method according to claim 4, characterized in that: The analyzing the vertical position and the horizontal position of the rail point cloud block to find the interference point cloud block representing the non-rail body specifically includes: Calculate the second vertical height of each data point of the rail point cloud block and its left adjacent point and right adjacent point; H ij1 =|z ij -z i(j+1) | H ij2 =|z ij -z i(j-1) |; Among them, Z ij is the height of the data point at the i-th position on the X-axis and the j-th position on the Y-axis, H ij1 H is the second vertical height between the data point and its right adjacent point. ij2 is the second vertical height between the data point and its left adjacent point; A second height threshold is set, and if the second vertical height is within the second height threshold, the left adjacent point or the right adjacent point is regarded as a pending ballast point cloud block; Importing historical data at the pending ballast point cloud block, and performing credibility evaluation on the pending ballast point cloud block; Among them, t is the time, k is the number of measurements, and Z tij is the height of the data point at position ij at time t, is the average height of the data points at position ij within the measurement time range, δ ij is the height standard deviation of the data points at position ij; If the data points in the pending ballast point cloud block meet the criteria for credibility determination, the pending ballast point cloud block is regarded as a ballast point cloud block, and the ballast point cloud block is excluded from the rail point cloud block; Wherein, μ is the error range, 2≤μ≤3; In the updated rail point cloud block, the horizontal distance between the edge data points in each horizontal height is calculated, and with reference to the rail design standard, the edge data points whose horizontal distance is greater than the rail design standard are extracted and regarded as pending foreign body point cloud blocks; Importing historical data of the undetermined foreign body point cloud block, and performing credibility assessment on the undetermined foreign body point cloud block; Among them, X tij is the vertical horizontal position of the data point at position ij at time t, is the average vertical and horizontal position of the data points at position ij within the measurement time range, Δ ij is the vertical horizontal position standard deviation of the data point at position ij; If the data points in the undetermined foreign body point cloud block meet the criteria for credibility determination, the undetermined foreign body point cloud block is regarded as a foreign body point cloud block; The ballast point cloud block and the foreign matter point cloud block are collectively regarded as interference point cloud blocks.

6. The rail fastener detection method according to claim 5, characterized in that: The method of scanning the track with a track inspection instrument to obtain a chord data map, and finding an interference point cloud block representing a fastener based on the chord data map, and using the interference point cloud block as a measurement point cloud block, specifically includes: Scanning the track with a track inspection instrument to obtain a chord data graph, and constructing a chord data matrix T according to the chord data graph; Among them, N is the total number of scanning and sampling times of the track inspection instrument, To scan the sampling data; Based on the matrix binary recursive construction method, the chord data matrix is ​​subjected to singular value decomposition to obtain similar singular values ​​and key singular values, and then similar signals ε and key signals λ corresponding to the similar singular values ​​and the key singular values ​​are obtained respectively; Decompose the similar signal level by level to obtain the similar signal ε at each level a and the key signal λ a ; where a is the number of decompositions; Construct a key signal matrix D based on the key signals at all levels a ; In the key signal matrix D a Find the signal of peak mutation in the , and regard the signal of peak mutation as the first pending signal; Analyze the signals with the same peaks and consistent spacing in the first pending signals, and regard them as the second pending signals; Referring to the fastener design specification, the second pending signal whose spacing meets the fastener design specification is regarded as a fastener signal, and the interference point cloud block at the fastener signal position is used as a measurement point cloud block.

7. The rail fastener detection method according to claim 6, characterized in that: The calculating the geometric position relationship between each data point in the measured point cloud block and the reference plane and the ground plane, and the geometric position relationship between adjacent data points, so as to determine the type of the fastener, specifically includes: Calculate the vertical distances between each data point in the measured point cloud block and the reference plane and the ground plane, respectively, and refer to the fastener design specification to regard the data points whose vertical distances meet the fastener design specification as the data points to be verified; H ij3 =z ij H ij4 =|z ij -z′ ij |; Among them, z′ ij is the height of the reference plane at the i-th position of the X axis and the j-th position of the Y axis, H ij3 is the vertical distance between the data point and the ground plane, H ij4 is the vertical distance between the data point and the reference plane; Calculate the slope between adjacent data points to be tested, and refer to the fastener design specification to find the corresponding fastener type; Among them, Yij is the horizontal position of the data point at position ij, G ij1 is the slope between the data point and the previous adjacent point, G ij2 is the slope between a data point and its adjacent point.

8. A rail fastener detection system, characterized in that: include: The first scanning module is used to scan the track in sequence using a laser radar to obtain a three-dimensional point cloud map of the track; Calculation module: used for considering the highest point of each position of the three-dimensional point cloud image as a reference plane to calculate the corresponding ground surface plane; The first analysis module is used to perform feature analysis on the three-dimensional point cloud image to find out the point cloud representing the rails to obtain a rail point cloud image; the rail point cloud image includes two rail point cloud blocks arranged relatively; The second analysis module is used to analyze the vertical position and horizontal position of the rail point cloud block to find out the interference point cloud block that does not represent the rail body; The second scanning module is used to scan the track using a track inspection instrument to obtain a chord data map, and based on the chord data map, find out the interference point cloud block representing the fastener and use it as the measurement point cloud block; Determination module: calculating the geometric position relationship between each data point in the measured point cloud block compared with the reference plane and the ground plane, and the geometric position relationship between adjacent data points, so as to determine the type of the fastener; Monitoring module: used for regularly monitoring the geometric position relationship between each data point in the measurement point cloud block compared with the reference plane and the ground surface plane, as well as the geometric position relationship between adjacent data points, so as to detect the integrity of each fastener.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the rail fastener detection method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the rail fastener detection method according to any one of claims 1 to 7 is implemented.

Images