Steel rail abrasion detection method and device based on full-section profile registration, electronic equipment and storage medium

Abstract

The invention discloses a steel rail abrasion detection method and device for full-section profile registration, electronic equipment and a storage medium, and the method comprises the steps: S1, carrying out the clustering filtering preprocessing of collected original profile point cloud data of a rail, and extracting the point cloud of a rail head region and the point cloud of a rail waist region according to the longitudinal amplitude feature of the point cloud and the profile curvature feature; s2, segmenting a point cloud characteristic arc of a rail waist area on the inner side of the steel rail profile to obtain a large arc and a small arc, calculating to obtain initial double circle centers of the segmented large arc and small arc, and fitting to obtain optimal double circle centers; s3, calculating inner side posture transformation parameters of the inner side profile according to the optimal double-circle-center positioning, and calculating outer side posture transformation parameters of the outer side profile of the steel rail through the inner side posture transformation parameters and system calibration parameters to realize full-section profile registration; and S4, calculating the abrasion value of the steel rail by utilizing the combination of the registered actually measured profile and the characteristic points on the inner side and the outer side of the standard profile of the steel rail. The method is high in detection precision, simple in calculation and good in detection real-time performance.

Description

Technical Field

[0001] The present application relates to the technical field of rail detection, and in particular, to a rail wear detection method, device, electronic equipment and storage medium with full-section profile registration. Background Art

[0002] As a fast and convenient means of transportation, subway has become an indispensable part of public travel. With the continuous increase in subway operating mileage, the safety and reliability of subway tracks have received increasing attention. The operation of multiple and high-frequency trains has aggravated the wear of rails, which directly affects operational safety. Therefore, real-time and efficient dynamic rail wear detection is of great significance to ensure subway operation safety and line maintenance.

[0003] At present, the common rail wear detection methods for subways include static contact detection and dynamic non-contact detection. Among them, static contact detection is to measure the rail wear value by manually using a wear caliper. The detection result is easily affected by human subjectivity. The long-term friction between the caliper and the rail will cause a certain degree of damage, affecting the accuracy and reliability of the measurement data; dynamic non-contact detection generally adopts a detection method based on a combination of array cameras and lasers. Its principle is mainly to collect the light image of the object being measured by the laser through the array camera, and obtain the actual point cloud coordinates of the measured rail profile through image processing and coordinate conversion calculation. After the profile is aligned by the algorithm, the rail wear is calculated. The detection equipment uses the vehicle as a carrier and moves with the vehicle to realize dynamic non-contact detection. Compared with the static contact detection method, dynamic non-contact detection has the advantages of no contact wear, fast and efficient, and high precision.

[0004] The focus of existing rail wear detection technical solutions is mainly on the extraction and alignment of two-dimensional feature point clouds of rail sections. Among them, the accurate extraction of the feature point clouds of the rail head and rail waist is an important prerequisite for profile alignment, but it is easily affected by external interference. Profile alignment is the key to ensuring the accuracy of wear detection. When extracting rail cross-section feature point clouds, existing technical solutions usually use single conditions such as profile curvature features and regional continuous convexity and concavity for extraction. In actual subway lines where multi-state interference factors coexist, there are problems such as susceptibility to interference, easy misjudgment, and poor real-time detection. When aligning rail profiles, existing technical solutions usually use ICP (Iterative Closest Point) and its improved method for profile alignment, which has defects such as sensitivity to initial values ​​and high calculation complexity. When calculating rail wear, existing technical solutions usually calculate the wear value through wear feature points according to standard definitions, without considering the offset error between the measured profile and the actual profile caused by track line interference factors (such as oil coating, steel letter printing, attached foreign matter, etc.) during dynamic detection, which will directly affect the wear detection accuracy.

[0005] In summary, the present invention proposes a wear detection method based on efficient registration of full-section rail profile to overcome the defects and shortcomings of existing rail wear detection technologies. Summary of the invention

[0006] On one hand, the present application provides a rail wear detection method with full-section profile registration to solve the technical problems of low detection accuracy, complex calculation and poor real-time detection in existing rail wear detection technologies.

[0007] This application is implemented through the following scheme:

[0008] A rail wear detection method with full cross-section profile registration comprises the following steps:

[0009] S1. Clustering and filtering pre-processing of the collected original track profile point cloud data is performed based on the distance-based clustering filtering algorithm. Then, the point cloud of the rail head and rail waist area is accurately extracted based on the method of combining the longitudinal amplitude characteristics of the point cloud with the profile curvature characteristics.

[0010] S2. Segment the characteristic arc of the point cloud of the inner rail waist area of ​​the rail profile to obtain large and small arcs, and use the perpendicular diameter theorem combined with the properties of similar triangles to first calculate a relatively accurate initial double center, and then use the nonlinear least squares algorithm based on radius constraints to fit the initial double center to obtain the optimal double center of the segmented large and small arcs;

[0011] S3, calculating the inner posture transformation parameters of the inner profile of the rail according to the optimal double center positioning obtained by fitting, calculating the outer posture transformation parameters of the outer profile of the rail by combining the inner posture transformation parameters with the system calibration parameters, and realizing the full-section profile registration of the measured profile and the standard profile;

[0012] S4. Calculate the rail wear value by combining the characteristic points on the inner and outer sides of the rail profile of the measured profile after profile registration with the standard profile.

[0013] Furthermore, before collecting the original track profile point cloud data, the following steps are also included:

[0014] S0. Calibrate the system parameters of the rail wear detection device composed of four line laser sensors installed on the detection beam to realize the conversion of the original coordinate system of each line laser to the unified coordinate system, and obtain the system calibration parameters, including the transformation rotation angle θ between the original coordinate system and the unified coordinate system, The origin of a coordinate system is in the X direction and the Z direction The relative offset Δ x , Δz, where the X direction is the rail The Z direction is the horizontal direction of the cross section, and the Z direction is the vertical direction of the rail cross section height.

[0015] Furthermore, the step S1 specifically includes the steps of:

[0016] S11. According to the distribution characteristics of the measured original silhouette point cloud, which has dense normal data and sparse outlier interference data, a distance-based clustering algorithm is used to segment the point cloud to obtain a segmented set S = {P i ={(x j , z j )}} i，j∈N , where i represents the segment point set index, j represents the index of the point in the single point set, and the search starts from the first point and stores it in the point set P i If the distance between the current point and the next point is less than or equal to the set threshold T1, the next point is stored in the point set P i ; If it is greater than the set threshold T1, the next point is stored in the new point set P i+1 , and so on, the search stops when all points are traversed;

[0017] S12, set the threshold T2, calculate the distance between the last point of the previous point set and the first point of the next point set in the set S, if the distance is less than the threshold T2, then merge the two point sets into a new point set to avoid segmenting the outline belonging to the same feature area into multiple point clouds, which affects the subsequent feature point cloud extraction;

[0018] S13, setting a threshold T3, if the number of points in a single point set in the set S is less than the threshold T3, the point set is filtered out to eliminate the outlier interference point cloud in the original silhouette;

[0019] S14. According to the actual collection profile point cloud characteristics, the rail head area point cloud contains the maximum longitudinal amplitude point, and the index h of the potential rail head area point set is calculated, and the corresponding point set P is quickly extracted from the set S. h ; The extracted rail head point set P is determined by the positional relationship between the rail head and the rail waist h Is it correct? The rail waist area point set index of the rail outer profile is smaller than the rail head area point set, and the inner profile of the rail is the opposite. If the rail head point set is extracted correctly, continue to execute step S15, and if it is wrong, jump to step S16;

[0020] S15. Due to the installation angle of the line laser, there is an obvious discontinuity feature between the rail head and the rail waist, and the rail head and rail waist point sets are adjacent. Therefore, the index w of the rail waist area point set is calculated, and the corresponding point set P is quickly extracted from the set S. w , and sequentially merge the rail head and rail waist point sets to obtain the effective profile point set P hw ;

[0021] S16. Based on the characteristics that both the rail head and the rail waist are curves, the curvature of each point set in the set S and the longitudinal amplitude of the first point of the point set are calculated, and the indexes h and w corresponding to the rail head and rail waist area point sets are respectively selected, and the corresponding rail head point set P is quickly extracted from the set S. h and the set of rail waist points Pw , and merge the point sets in order to obtain the effective contour point set P hw If the rail head and rail waist area point sets are not screened out, it is judged as an invalid profile and discarded.

[0022] Furthermore, the step S2 specifically includes the steps of:

[0023] S21, perform point cloud segmentation on two large and small arcs with radii R350 and R20 in the inner waist area of ​​the standard 50kg / m rail profile to obtain a large arc point set P R350 and the small arc point set P R20 ;

[0024] S22. Fit the two divided arcs to obtain the optimal double center. When fitting, first use the perpendicular diameter theorem combined with the properties of similar triangles to calculate a relatively accurate initial double center, and then use the nonlinear least squares method based on radius constraints to fit the optimal double center.

[0025] Furthermore, step S22 specifically includes the steps of:

[0026] S221, Great Circle Point Set P R350 In the example, we randomly select two points E and F, set the origin of the arc point cloud coordinate system as O, the midpoint of the chord EF as G, the center of the fitted circle as H, and the line segment GH as the perpendicular bisector of the chord EF. Calculate the vector and vector Model:

[0027]

[0028] in is the radius of the arc, and the vector is obtained based on the properties of similar triangles

[0029]

[0030] According to the known conditions, the center H is obtained:

[0031]

[0032] Assume that the coordinates of the fitting circle center H are (x c ,z c ), calculate the point set P R350 ={(x i ,z i )} i∈N The distance from each point to the center of the fitted circle H and the radius of the circle R 350 The cumulative sum of the differences is taken as the fitting center error, and the fitting error is set to e:

[0033]

[0034] Where n represents the number of points in the point set;

[0035] According to the above method, set the threshold t = 20, repeat the calculation t times, and store the fitting error in the set S e ={e t} t∈N , find the set S e The fitting circle center H corresponding to the minimum fitting error;

[0036] S222, using the least squares nonlinear fitting Levenberg-Marquardt algorithm based on radius constraints to quickly iterate and calculate the optimal circle center, and transform the point set P R350 ={(x i ,z i )} i∈N , circle radius R 350 , the center of the circle H(x c ,z c ) is used as the input parameter of the algorithm iteration and the target residual function is defined as:

[0037]

[0038] Where M represents the number of fitting points. When F(x) takes the minimum value, the optimal center Q of the large arc is obtained. R350 ;

[0039] S223, in the small arc point set P R20 In the example, we randomly select two points E and F, set the origin of the arc point cloud coordinate system as O, the midpoint of the chord EF as G, the center of the fitted circle as H, and the line segment GH as the perpendicular bisector of the chord EF. Calculate the vector and vector Model:

[0040]

[0041] in is the radius of the arc, and the vector is obtained based on the properties of similar triangles

[0042]

[0043] According to the known conditions, the center H is obtained:

[0044]

[0045] Assume that the coordinates of the fitting circle center H are (x c ,z c ), calculate the point set P R20 ={(x i ,z i )} i∈NThe distance from each point to the center of the fitted circle H and the radius of the circle R 20 The cumulative sum of the differences is taken as the fitting center error, and the fitting error is set to e:

[0046]

[0047] Where n represents the number of points in the point set;

[0048] According to the above method, set the threshold t = 20, repeat the calculation t times, and store the fitting error in the set S e ={e t} t∈N , find the set S e The fitting circle center H corresponding to the minimum fitting error;

[0049] S224, using the least squares nonlinear fitting Levenberg-Marquardt algorithm based on radius constraints to quickly iterate and calculate the optimal circle center, and convert the point set P R20 ={(x i ,z i )} i∈N , circle radius R 20 , the center of the circle H(x c ,z c ) is used as the input parameter of the algorithm iteration and the target residual function is defined as:

[0050]

[0051] Where M represents the number of fitting points. When F(x) takes the minimum value, the optimal center Q of the small arc is obtained. R20 .

[0052] Furthermore, the step S3 specifically includes the steps of:

[0053] S31, assuming that the optimal center coordinates of the large and small arcs are (Cx 350 ,Cz 350 ) and (Cx 20 ,Cz 20 ), the coordinates of the centers of the large and small arcs at the corresponding positions of the standard 50kg / m rail profile are (Cx B350 ,Cz B350 ) and (Cx B20 ,Cz B20 ), calculate the rotation angle θ, rotation matrix R, and translation matrix T for the registration of the original profile and the standard profile, and complete the registration of the inner profile:

[0054]

[0055] S32, assuming that the medial profile registration rotation and translation matrices are R I, T I , the inner profile calibration rotation matrix is ​​R Ibd , the rotation and translation matrices of the inner profile from the unified coordinate system and the standard profile coordinate system are obtained as R bz , T bz :

[0056] R bz =R I *R Ibd T

[0057] T bz =T I

[0058] Assume that the calibration rotation and translation matrices of the outer contour are R Obd , T Obd , before and after the registration, a certain point is (x i ,z i ) and (x i ',z i '), thus completing the outer contour registration:

[0059]

[0060] At this point, the full cross-section profile alignment of the inner and outer sides of the rail profile is completed.

[0061] Furthermore, the step S4 specifically includes the steps of:

[0062] S41. Measure the characteristic points on the inner and outer sides of the rail profile of the measured profile and the standard profile after registration, including the coordinates of the inner vertical wear points of the standard profile and the measured profile, respectively. bv ,z bv )、V'(x v ,z v ), the coordinates of the inner and outer rail jaw points of the standard profile are A in (x bie ,z bie ), A out (x boe ,z boe ), the coordinates of the inner and outer rail jaw points of the measured profile are A i ' n (x ie ,z ie ), A' out (x oe ,z oe ), the coordinates of the wear points on the inner and outer sides of the standard profile are H in (x bih ,z bih ), H out (xboh ,z boh ), the coordinates of the wear points on the inner and outer sides of the measured profile are H i ' n (x ih ,z ih ), H' out (x oh ,z oh );

[0063] S42. Calculate the vertical wear of the rail based on the characteristic points on the inner and outer sides of the rail profile. v 、Side wearω h and total wear ω a :

[0064]

[0065] On the other hand, the present application also provides a rail wear detection device with full cross-section profile registration, comprising:

[0066] The point cloud preprocessing and feature extraction module is used to cluster and filter the collected original track profile point cloud data based on the distance-based clustering and filtering algorithm, and then accurately extract the point cloud of the rail head and rail waist area based on the method of combining the longitudinal amplitude characteristics of the point cloud with the profile curvature characteristics;

[0067] The optimal double center fitting module is used to segment the point cloud feature arcs in the inner rail waist area of ​​the rail profile to obtain large and small arcs, and to obtain the optimal double centers of the segmented large and small arcs through the nonlinear least squares algorithm based on radius constraints;

[0068] The full-section profile registration module is used to calculate the inner side posture transformation parameters of the inner side profile of the rail according to the optimal dual-center positioning obtained by fitting, and calculate the outer side posture transformation parameters of the outer side profile of the rail by combining the inner side posture transformation parameters with the system calibration parameters, so as to realize the full-section profile registration of the measured profile and the standard profile;

[0069] The rail wear value calculation module is used to calculate the rail wear value by combining the measured profile after profile registration with the characteristic points on the inner and outer sides of the rail profile of the standard profile.

[0070] On the other hand, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the rail wear detection method with full-section profile registration when executing the computer program.

[0071] On the other hand, the present application further provides a storage medium, which includes a stored program, and when the program is run, controls the device where the storage medium is located to execute the steps of the rail wear detection method with full-section profile registration.

[0072] On the other hand, the present application also provides a computer program product, including a computer program or computer executable instructions, which, when executed by a processor, implement the steps of the rail wear detection method with full-section profile registration as described above.

[0073] Compared with the prior art, this application has the following beneficial effects:

[0074] (1) When processing the collected original track profile point cloud data, the present application adopts a distance-based clustering and filtering algorithm to effectively filter out interfering outlier point clouds, making the extracted rail profile data more accurate.

[0075] (2) This application uses a feature point cloud segmentation algorithm that combines the longitudinal amplitude features of the point cloud with the profile curvature features to accurately extract and segment the point clouds of the rail head and rail waist areas, ensuring higher accuracy in profile matching.

[0076] (3) This application uses the perpendicular diameter theorem combined with the properties of similar triangles to first calculate a relatively accurate initial double center of the circle, and then uses the nonlinear least squares method based on radius constraints to fit the optimal double center of the circle, thereby reducing the algorithm's sensitivity to the initial value, improving the algorithm's iteration efficiency, and improving the system's real-time detection performance.

[0077] (4) The inner side of the rail profile of the present application is aligned by double-center positioning through rail waist fitting, and the outer side of the profile is aligned by combining the inner side posture transformation parameters with the system calibration parameters, thereby achieving rapid alignment of the full-section profile and improving the real-time performance of the system detection.

[0078] (5) The present application uses a combination of characteristic points on both sides of the rail to calculate the wear value, which effectively eliminates the offset error caused by interference during the dynamic detection process, and the wear detection accuracy is higher under multi-state working conditions.

[0079] In addition to the above-described purposes, features and advantages, the present application also has other purposes, features and advantages. The present application will be further described in detail with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application.

[0081] Figure 1 It is a schematic flow chart of a rail wear detection method with full-section profile registration according to a preferred embodiment of the present application.

[0082] Figure 2 It is a schematic structural diagram of a rail wear detection device according to a preferred embodiment of the present application.

[0083] Figure 3 This is a schematic diagram of the standard 50kg / m rail profile.

[0084] Figure 4 This is the effect picture before full section profile registration.

[0085] Figure 5 This is the effect picture after full section profile registration.

[0086] Figure 6 It is a schematic diagram of wear calculation according to standard definition.

[0087] Figure 7 It is a schematic diagram of wear calculation when the profile is offset.

[0088] Figure 8 This is a schematic diagram of a rail wear detection device module for full cross-section profile registration according to a preferred embodiment of the present application. picture.

[0089] Fig. 9 It is a schematic diagram of a module of a rail wear detection device with full-section profile registration according to a preferred embodiment of the present application.

[0090] Fig.10 It is a schematic block diagram of an electronic device entity of a preferred embodiment of the present application.

[0091] Fig.11 It is a diagram of the internal structure of a computer device of a preferred embodiment of the present application.

[0092] In the figure: 1. Detection beam; 2. Line laser. DETAILED DESCRIPTION

[0093] The embodiments of the present application are described in detail below in conjunction with the accompanying drawings, but the present application can be implemented in a variety of different ways defined and covered below.

[0094] like Figure 1 As shown, the preferred embodiment of the present application provides a rail wear detection method with full-section profile registration, comprising the steps of:

[0095] S1. Clustering and filtering pre-processing of the collected original track profile point cloud data is performed based on the distance-based clustering filtering algorithm. Then, the point cloud of the rail head and rail waist area is accurately extracted based on the method of combining the longitudinal amplitude characteristics of the point cloud with the profile curvature characteristics.

[0096] S2. Segment the characteristic arc of the point cloud of the inner rail waist area of ​​the rail profile to obtain large and small arcs, and use the perpendicular diameter theorem combined with the properties of similar triangles to first calculate a relatively accurate initial double center, and then use the nonlinear least squares algorithm based on radius constraints to fit the initial double center to obtain the optimal double center of the segmented large and small arcs;

[0097] S3, calculating the inner posture transformation parameters of the inner profile of the rail according to the optimal double center positioning obtained by fitting, calculating the outer posture transformation parameters of the outer profile of the rail by combining the inner posture transformation parameters with the system calibration parameters, and realizing the full-section profile registration of the measured profile and the standard profile;

[0098] S4. Calculate the rail wear value by combining the characteristic points on the inner and outer sides of the rail profile of the measured profile after profile registration with the standard profile.

[0099] Compared with the prior art, this embodiment has the following beneficial effects:

[0100] (1) This embodiment uses a distance-based clustering and filtering algorithm to process the collected original track profile point cloud data, effectively filtering out interfering outlier point clouds, making the extracted rail profile data more accurate. Yes;

[0101] (2) This embodiment uses a feature point cloud segmentation algorithm that combines the longitudinal amplitude feature of the point cloud with the profile curvature feature to accurately extract and segment the point clouds of the rail head and rail waist areas, ensuring a higher accuracy of profile matching;

[0102] (3) This embodiment uses the perpendicular diameter theorem combined with the properties of similar triangles to first calculate a relatively accurate initial double center, and then uses the nonlinear least squares method based on radius constraints to fit the optimal double center, thereby reducing the algorithm's sensitivity to the initial value, improving the algorithm's iteration efficiency, and improving the system's real-time detection performance;

[0103] (4) The inner side of the rail profile of the present application is positioned and registered by rail waist fitting double center, and the outer side of the profile is registered by combining the inner side posture transformation parameters with the system calibration parameters, so as to achieve rapid registration of the full cross-section profile and improve the real-time performance of the system detection;

[0104] (5) This embodiment uses a combination of characteristic points on both sides of the rail to calculate the wear value, which effectively eliminates the offset error caused by interference during the dynamic detection process, and the wear detection accuracy is higher under multi-state working conditions.

[0105] Preferably, before collecting the original track profile point cloud data, the method further includes the following steps:

[0106] S0, four line laser sensors are installed on the detection beam 1 to form a rail wear detection device (such as Figure 2 As shown), the system parameter calibration is completed, and the conversion of the original coordinate system of each line laser 2 to the unified coordinate system is realized, and the system calibration parameters are obtained, including the transformation rotation angle θ between the original coordinate system and the unified coordinate system, and the relative offset Δx and Δz of the origin of the original coordinate system and the unified coordinate system in the X direction and the Z direction, wherein the X direction is the horizontal direction of the cross section of the rail, and the Z direction is the vertical direction of the height of the cross section of the rail.

[0107] In this embodiment, the rail wear detection device includes a detection beam and four line lasers 2. The rail inspection beam is rigidly connected to the bottom of the engineering vehicle, and the line lasers 2 are fixedly installed under the detection beam at a certain angle. The object to be measured is a 50kg / m rail as an example. The side where the rail and the wheel rub is defined as the inner side, and the non-friction side is defined as the outer side. The detection device is mounted on the track inspection engineering vehicle, and the engineering vehicle needs to be parked on a straight track and calibrated when it is in a stationary state. During calibration, the line laser sensor is turned on, the customized calibration rod is placed on the track surface, and the calibration rod is moved so that the laser line coincides with the positioning mark on the calibration rod for positioning and alignment; calibration is performed through a software calibration program, mainly realizing the conversion of the original coordinate system of each line laser to a unified coordinate system, and calculating the output calibration parameters including the rotation angle θ of the two coordinate systems, and the relative offsets of the origins of the two coordinate systems in the X direction (horizontal direction of the cross section) and the Z direction (vertical direction of height) are Δx and Δz, respectively, so as to provide necessary preliminary preparations for subsequent detection.

[0108] Preferably, the step S1 specifically comprises the steps of:

[0109] S11. According to the distribution characteristics of the measured original silhouette point cloud, which has dense normal data and sparse outlier interference data, a distance-based clustering algorithm is used to segment the point cloud to obtain a segmented set S = {P i ={(x j , z j )}} i，j∈N , where i represents the segment point set index, j represents the index of the point in the single point set, and the search starts from the first point and stores it in the point set P i If the distance between the current point and the next point is less than or equal to the set threshold T1, the next point is stored in the point set P i ; If it is greater than the set threshold T1, the next point is stored in the new point set P i+1 , and so on, the search stops when all points are traversed;

[0110] S12, set the threshold T2, calculate the distance between the last point of the previous point set and the first point of the next point set in the set S, if the distance is less than the threshold T2, then merge the two point sets into a new point set to avoid segmenting the outline belonging to the same feature area into multiple point clouds, which affects the subsequent feature point cloud extraction;

[0111] S13, setting a threshold T3, if the number of points in a single point set in the set S is less than the threshold T3, the point set is filtered out to eliminate the outlier interference point cloud in the original silhouette;

[0112] S14. According to the actual collection profile point cloud characteristics, the rail head area point cloud contains the maximum longitudinal amplitude point, and the index h of the potential rail head area point set is calculated, and the corresponding point set P is quickly extracted from the set S. h; The extracted rail head point set P is determined by the positional relationship between the rail head and the rail waist h Is it correct? The rail waist area point set index of the rail outer profile is smaller than the rail head area point set, and the inner profile of the rail is the opposite. If the rail head point set is extracted correctly, continue to execute step S15, and if it is wrong, jump to step S16;

[0113] S15. Due to the installation angle of the line laser, there is an obvious discontinuity feature between the rail head and the rail waist, and the rail head and rail waist point sets are adjacent. Therefore, the index w of the rail waist area point set is calculated, and the corresponding point set P is quickly extracted from the set S. w , and sequentially merge the rail head and rail waist point sets to obtain the effective profile point set P hw ;

[0114] S16. Based on the characteristics that both the rail head and the rail waist are curves, the curvature of each point set in the set S and the longitudinal amplitude of the first point of the point set are calculated, and the indexes h and w corresponding to the rail head and rail waist area point sets are respectively selected, and the corresponding rail head point set P is quickly extracted from the set S. h and the set of rail waist points P w , and merge the point sets in order to obtain the effective contour point set P hw If the rail head and rail waist area point sets are not screened out, it is judged as an invalid profile and discarded.

[0115] Affected by the polymorphic interference factors of the actual track line, such as reflection in strong light areas, roadbed, fasteners, foreign objects on the railside, etc., there are outlier interference points in the original rail profile point cloud collected by line laser, which will affect the profile alignment and wear calculation accuracy. Therefore, this embodiment pre-processes the original profile point cloud to eliminate outlier interference points, so that the extracted rail profile data is more accurate, and at the same time, the effective profile is quickly identified and the point cloud of the rail head and rail waist area is extracted, and the point cloud of the rail head and rail waist area is accurately extracted and segmented to ensure higher accuracy of profile matching.

[0116] Preferably, the step S2 specifically comprises the steps of:

[0117] S21, perform point cloud segmentation on two large and small arcs with radii R350 and R20 in the inner waist area of ​​the standard 50kg / m rail profile to obtain a large arc point set P R350 and the small arc point set P R20 ;

[0118] S22. Fit the two divided arcs to obtain the optimal double center. When fitting, first use the perpendicular diameter theorem combined with the properties of similar triangles to calculate a relatively accurate initial double center, and then use the nonlinear least squares method based on radius constraints to fit the optimal double center.

[0119] Figure 3This is a schematic diagram of the standard 50kg / m rail profile. It is necessary to segment the point cloud of two large and small arcs with radii R350 and R20 in the rail waist area inside the rail profile, and extract the arcs as the registration feature point cloud. Among them, point A is the rail jaw point, the R350 large arc is the BC part in the figure, and the R20 small arc is the CD part in the figure. The two arcs are tangent at point C.

[0120] According to the standard 50kg / m rail size parameters, the distance between points AC and AD can be accurately calculated, and thus the effective profile point set P can be obtained. hw The two arc point sets are accurately segmented and extracted as P R350 and P R20 .

[0121] The optimal double center is obtained by fitting the divided large and small arcs. The circle perpendicular diameter theorem combined with the properties of similar triangles is first used to calculate a relatively accurate initial double center, and then the nonlinear least squares method based on radius constraint is used to fit the optimal double center.

[0122] This embodiment uses the perpendicular diameter theorem combined with the properties of similar triangles to first calculate a relatively accurate initial double center of the inner waist area of ​​the standard 50kg / m rail profile, and then uses the nonlinear least squares method based on radius constraints to fit the optimal double center of the two large and small arcs, thereby reducing the algorithm's sensitivity to the initial value, improving the algorithm's iteration efficiency, and improving the system's real-time detection performance.

[0123] Specifically, step S22 specifically includes the steps of:

[0124] S221, Great Circle Point Set P R350 In the example, we randomly select two points E and F, set the origin of the arc point cloud coordinate system as O, the midpoint of the chord EF as G, the center of the fitted circle as H, and the line segment GH as the perpendicular bisector of the chord EF. Calculate the vector and vector Model:

[0125]

[0126] in is the radius of the arc, and the vector is obtained based on the properties of similar triangles

[0127]

[0128] According to the known conditions, the center H is obtained:

[0129]

[0130] Assume that the coordinates of the fitting circle center H are (x c ,z c ), calculate the point set PR350 ={(x i ,z i )} i∈N The distance from each point to the center of the fitted circle H and the radius of the circle R 350 The cumulative sum of the differences is taken as the fitting center error, and the fitting error is set to e:

[0131]

[0132] Where n represents the number of points in the point set;

[0133] According to the above method, set the threshold t = 20, repeat the calculation t times, and store the fitting error in the set S e ={e t} t∈N , find the set S e The fitting circle center H corresponding to the minimum fitting error;

[0134] S222, using the least squares nonlinear fitting Levenberg-Marquardt algorithm based on radius constraints to quickly iterate and calculate the optimal circle center, and transform the point set P R350 ={(x i ,z i )} i∈N , circle radius R 350 , the center of the circle H(x c ,z c ) is used as the input parameter of the algorithm iteration and the target residual function is defined as:

[0135]

[0136] Where M represents the number of fitting points. When F(x) takes the minimum value, the optimal center Q of the large arc is obtained. R350 ;

[0137] S223, in the small arc point set P R20 In the example, we randomly select two points E and F, set the origin of the arc point cloud coordinate system as O, the midpoint of the chord EF as G, the center of the fitted circle as H, and the line segment GH as the perpendicular bisector of the chord EF. Calculate the vector and vector Model:

[0138]

[0139] in is the radius of the arc, and the vector is obtained based on the properties of similar triangles

[0140]

[0141] According to the known conditions, the center H is obtained:

[0142]

[0143] Assume that the coordinates of the fitting circle center H are (x c ,z c ), calculate the point set P R20 ={(x i ,z i )} i∈N The distance from each point to the center of the fitted circle H and the radius of the circle R 20 The cumulative sum of the differences is taken as the fitting center error, and the fitting error is set to e:

[0144]

[0145] Where n represents the number of points in the point set;

[0146] According to the above method, set the threshold t = 20, repeat the calculation t times, and store the fitting error in the set S e ={e t} t∈N , find the set S e The fitting circle center H corresponding to the minimum fitting error;

[0147] S224, using the least squares nonlinear fitting Levenberg-Marquardt algorithm based on radius constraints to quickly iterate and calculate the optimal circle center, and convert the point set P R20 ={(x i ,z i )} i∈N , circle radius R 20 , the center of the circle H(x c ,z c ) is used as the input parameter of the algorithm iteration and the target residual function is defined as:

[0148]

[0149] Where M represents the number of fitting points. When F(x) takes the minimum value, the optimal center Q of the small arc is obtained. R20 .

[0150] This embodiment provides a method of using the perpendicular diameter theorem combined with the properties of similar triangles to first calculate a relatively accurate initial double center of the inner waist area of ​​the standard 50kg / m rail profile, and then use the nonlinear least squares method based on radius constraints to fit the optimal double center of two large and small arcs, thereby reducing the algorithm's sensitivity to initial values, improving the algorithm's iteration efficiency, and improving the system's real-time detection performance.

[0151] Preferably, the step S3 specifically comprises the steps of:

[0152] S31, assuming that the optimal double center coordinates of the large and small arc fitting are (Cx 350 ,Cz 350 ) and (Cx 20 ,Cz 20 ), the coordinates of the centers of the large and small arcs at the corresponding positions of the standard 50kg / m rail profile are (Cx B350 ,Cz B350 ) and (Cx B20 ,Cz B20 ), calculate the rotation angle θ, rotation matrix R, and translation matrix T for the registration of the original profile and the standard profile, and complete the registration of the inner profile:

[0153]

[0154] S32, assuming that the medial profile registration rotation and translation matrices are R I , T I , the inner profile calibration rotation matrix is ​​R Ibd , the rotation and translation matrices of the inner profile from the unified coordinate system and the standard profile coordinate system are obtained as R bz , T bz :

[0155] R bz =R I *R Ibd T

[0156] T bz =T I

[0157] Assume that the calibration rotation and translation matrices of the outer contour are R Obd , T Obd , before and after the registration, a certain point is (x i ,z i ) and (x i ',z i '), thus completing the outer contour registration:

[0158]

[0159] At this point, the full-section profile registration of the inner and outer sides of the rail profile is completed. The effects before and after the full-section profile registration are as follows: Figure 4 and Figure 5 shown.

[0160] The full-section profile registration of the rail in this embodiment includes the registration of the inner side (wear side) and the outer side (non-wear side). The inner side is registered by positioning the double center of the rail waist, and the outer side is registered by combining the inner side posture transformation parameters with the system calibration parameters to achieve rapid registration of the full-section profile. Among them, the outer side profile registration involves three coordinate systems, namely the original profile coordinate system, the unified coordinate system, and the standard profile coordinate system. The outer side profile registration is to convert the outer side profile point cloud from the original profile coordinate system to the standard profile coordinate system, that is:

[0161] First, the conversion matrix between the inner and outer original profile coordinate systems and the unified coordinate system can be obtained through system calibration, namely the calibration matrix;

[0162] Then, according to the obtained medial profile registration matrix and the known medial profile calibration matrix, the transformation matrix between the profile from the unified coordinate system and the standard profile coordinate system can be obtained;

[0163] Finally, the original profile of the outer side of the rail is aligned with the profile through the above transformation relationship.

[0164] In addition, for elevated bridge track lines, guardrails will be laid on the inside of the track to prevent train derailment. Since the guardrail is close to the line rails, the feature point cloud of the inner side profile of the rails is blocked and missing. At this time, it can automatically switch to prioritize the use of the outer side profile point cloud according to step S31 to perform inner side profile alignment. On this basis, the inner side profile point cloud is used to perform outer side profile alignment according to step S32, thereby completing the full cross-section profile alignment of the rail.

[0165] Preferably, the step S4 specifically comprises the steps of:

[0166] S41. Measure the characteristic points on the inner and outer sides of the rail profile of the measured profile and the standard profile after registration, including the coordinates of the inner vertical wear points of the standard profile and the measured profile, respectively. bv ,z bv )、V'(x v ,z v ), the coordinates of the inner and outer rail jaw points of the standard profile are A in (x bie ,z bie ), A out (x boe ,z boe ), the coordinates of the inner and outer rail jaw points of the measured profile are A i ' n (x ie ,z ie ), A' out (x oe ,z oe ), the coordinates of the wear points on the inner and outer sides of the standard profile are H in (xbih ,z bih ), H out (x boh ,z boh ), the coordinates of the wear points on the inner and outer sides of the measured profile are H i ' n (x ih ,z ih ), H' out (x oh ,z oh );

[0167] S42. Calculate the vertical wear of the rail based on the characteristic points on the inner and outer sides of the rail profile. v 、Side wearω h and total wear ω a :

[0168]

[0169] Figure 6 It is a schematic diagram of wear calculation defined by the standard. Vertical wear is defined as the thickness of the rail tread worn in the vertical direction at 1 / 3 of the rail head width from the working edge; side wear is defined as the width of the working edge worn in the horizontal direction 16mm below the top of the rail; total wear is defined as the sum of vertical wear and 1 / 2 side wear.

[0170] In this embodiment, the deviation error between the measured profile and the real profile caused by the interference factor of the track line during the dynamic detection process will directly affect the wear detection accuracy. Figure 7 ) to calculate the wear value in combination to ensure the accuracy of wear detection, specifically:

[0171] Calculate the difference between the distance between the side wear points on the inside and outside of the measured profile and the distance between the side wear points on the sides of the standard profile, which is the side wear value; calculate the rail jaw points and inner vertical wear points on the inside and outside of the measured profile. The rail jaws will not be worn. The inner vertical grinding point to the inner rail jaw point corresponding to the measured profile and the standard profile The height difference is taken as the vertical wear value.

[0172] Since this embodiment uses a combination of multiple characteristic points on both sides of the rail to calculate vertical wear, side wear and total wear, it effectively eliminates the offset error caused by interference during dynamic detection, and the wear detection accuracy is higher under multi-state working conditions.

[0173] like Figure 8 As shown, another preferred embodiment of the present application further provides a rail wear detection device with full cross-section profile registration, comprising:

[0174] The point cloud preprocessing and feature extraction module is used to cluster and filter the collected original track profile point cloud data based on the distance-based clustering and filtering algorithm, and then accurately extract the point cloud of the rail head and rail waist area based on the method of combining the longitudinal amplitude characteristics of the point cloud with the profile curvature characteristics;

[0175] The optimal double center fitting module is used to segment the point cloud feature arcs in the inner rail waist area of ​​the rail profile to obtain large and small arcs, and to obtain the optimal double centers of the segmented large and small arcs through the nonlinear least squares algorithm based on radius constraints;

[0176] The full-section profile registration module is used to calculate the inner side posture transformation parameters of the inner side profile of the rail according to the optimal dual-center positioning obtained by fitting, and calculate the outer side posture transformation parameters of the outer side profile of the rail by combining the inner side posture transformation parameters with the system calibration parameters, so as to realize the full-section profile registration of the measured profile and the standard profile;

[0177] The rail wear value calculation module is used to calculate the rail wear value by combining the measured profile after profile registration with the characteristic points on the inner and outer sides of the rail profile of the standard profile.

[0178] Preferably, if Fig. 9 As shown, the rail wear detection device with full-section profile registration also includes:

[0179] The system parameter calibration module is used to calibrate the system parameters of the rail wear detection device composed of four line laser sensors installed on the detection beam, realize the conversion of the original coordinate system of each line laser to the unified coordinate system, and obtain the system calibration parameters, including the transformation rotation angle θ between the original coordinate system and the unified coordinate system, and the relative offset Δx and Δz between the origin of the original coordinate system and the unified coordinate system in the X direction and the Z direction, where the X direction is the horizontal direction of the rail cross section, and the Z direction is the vertical direction of the rail cross section height.

[0180] like Fig.10 As shown, the preferred embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes The computer program is implemented The rail wear detection method of the full cross-section profile registration in the above embodiment is step.

[0181] The preferred embodiment of the present application also provides a computer device, which can be a terminal or a liveness detection server, and its internal structure diagram can be as follows: Fig.11As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with other external computer devices through a network connection. When the computer program is executed by the processor, the steps of the rail wear detection method for full-section profile registration are implemented.

[0182] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0183] A preferred embodiment of the present application further provides a storage medium, which includes a stored program, and when the program is executed, controls the device where the storage medium is located to execute the steps of the rail wear detection method with full-section profile registration in the above embodiment.

[0184] The preferred embodiment of the present application provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of the electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the rail wear detection method for full-section profile registration described in the embodiment of the present application.

[0185] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0186] If the functions described in the method of this embodiment are implemented in the form of software functional units and sold or used as independent products, they can be stored in one or more computing device readable storage media. The understanding of this application The part of the embodiment that contributes to the prior art or the part of the technical solution can beIt is embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computing device (which may be a personal computer, a server, a mobile computing device or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0187] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0188] 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 the 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 generate 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.

[0189] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 A function specified in one or more boxes.

[0190] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented processing. management, so that in the computer or Instructions executed on other programmable devices are provided to implement Figure 1 indivual Process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0191] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0192] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A rail wear detection method with full cross-section profile registration, characterized in that: Includes steps: S1. Clustering and filtering pre-processing of the collected original track profile point cloud data is performed based on the distance-based clustering filtering algorithm. Then, the point cloud of the rail head and rail waist area is accurately extracted based on the method of combining the longitudinal amplitude characteristics of the point cloud with the profile curvature characteristics. S2. Segment the characteristic arc of the point cloud of the inner rail waist area of ​​the rail profile to obtain large and small arcs, and first calculate the initial double center by using the circle perpendicular diameter theorem combined with the properties of similar triangles. Then, the optimal double center of the large and small arcs after segmentation is obtained by fitting based on the initial double center through a nonlinear least squares algorithm based on radius constraints; S3, calculating the inner posture transformation parameters of the inner profile of the rail according to the optimal double center positioning obtained by fitting, calculating the outer posture transformation parameters of the outer profile of the rail by combining the inner posture transformation parameters with the system calibration parameters, and realizing the full-section profile registration of the measured profile and the standard profile; S4. Calculate the rail wear value by combining the characteristic points on the inner and outer sides of the rail profile of the measured profile after profile registration with the standard profile.

2. The rail wear detection method based on full-section profile registration according to claim 1, characterized in that: Before collecting the original track profile point cloud data, the following steps are also included: S0. Calibrate the system parameters of the rail wear detection device composed of four line laser sensors installed on the detection beam to realize the conversion of the original coordinate system of each line laser to the unified coordinate system, and obtain the system calibration parameters, including the transformation rotation angle θ between the original coordinate system and the unified coordinate system, and the relative offset Δx and Δz between the origin of the original coordinate system and the unified coordinate system in the X direction and the Z direction, where the X direction is the horizontal direction of the rail cross section, and the Z direction is the vertical direction of the rail cross section height.

3. The rail wear detection method based on full-section profile registration according to claim 1, characterized in that: The step S1 specifically includes the following steps: S11. According to the distribution characteristics of the measured original silhouette point cloud, which has dense normal data and sparse outlier interference data, a distance-based clustering algorithm is used to segment the point cloud to obtain a segmented set S = {P i ={(x j , z j )}} i，j∈N , where i represents the segment point set index, j represents the index of the point in the single point set, and the search starts from the first point and stores it in the point set P i If the distance between the current point and the next point is less than or equal to the set threshold T1, the next point is stored in the point set P i ; If it is greater than the set threshold T1, the next point is stored in the new point set P i+1 , and so on, the search stops when all points are traversed; S12, set the threshold T2, calculate the distance between the last point of the previous point set and the first point of the next point set in the set S, if the distance is less than the threshold T2, then merge the two point sets into a new point set to avoid segmenting the outline belonging to the same feature area into multiple point clouds, which affects the subsequent feature point cloud extraction; S13, setting a threshold T3, if the number of points in a single point set in the set S is less than the threshold T3, the point set is filtered out to eliminate the outlier interference point cloud in the original silhouette; S14. According to the actual collection profile point cloud characteristics, the rail head area point cloud contains the maximum longitudinal amplitude point, and the index h of the potential rail head area point set is calculated, and the corresponding point set P is quickly extracted from the set S. h ; The extracted rail head point set P is determined by the positional relationship between the rail head and the rail waist h Is it correct? The rail waist area point set index of the rail outer profile is smaller than the rail head area point set, and the inner profile of the rail is the opposite. If the rail head point set is extracted correctly, continue to execute step S15, and if it is wrong, jump to step S16; S15. Due to the installation angle of the line laser, there is an obvious discontinuity feature between the rail head and the rail waist, and the rail head and rail waist point sets are adjacent. Therefore, the index w of the rail waist area point set is calculated, and the corresponding point set P is quickly extracted from the set S. w , and sequentially merge the rail head and rail waist point sets to obtain the effective profile point set P hw ; S16. Based on the characteristics that both the rail head and the rail waist are curves, the curvature of each point set in the set S and the longitudinal amplitude of the first point of the point set are calculated, and the indexes h and w corresponding to the rail head and rail waist point sets are respectively selected, and the corresponding rail head point set P is quickly extracted from the set S. h and the set of rail waist points P w , and merge the point sets in order to obtain the effective contour point set P hw If the rail head and rail waist area point sets are not screened out, it is judged as an invalid profile and discarded.

4. The rail wear detection method based on full-section profile registration according to claim 1, characterized in that: The step S2 specifically includes the following steps: S21, perform point cloud segmentation on two large and small arcs with radii R350 and R20 in the inner waist area of ​​the standard 50kg / m rail profile to obtain a large arc point set P R350 and the small arc point set P R20 ; S22. Fit the two divided arcs to obtain the optimal double center. When fitting, first use the perpendicular diameter theorem combined with the properties of similar triangles to calculate the initial double center, and then use the nonlinear least squares method based on radius constraint to fit the optimal double center.

5. The rail wear detection method based on full cross-section profile registration according to claim 4, characterized in that: Step S22 specifically includes the following steps: S221, Great Circle Point Set P R350 In the example, we randomly select two points E and F, set the origin of the arc point cloud coordinate system as O, the midpoint of the chord EF as G, the center of the fitting circle as H, and the line segment GH as the perpendicular bisector of the chord EF. Calculate the vector and vector Model: in is the radius of the arc, and the vector is obtained based on the properties of similar triangles According to the known conditions, the center H is obtained: Assume that the coordinates of the fitting circle center H are (x c ,z c ), calculate the point set P R350 ={(x i ,z i )} i∈N The distance from each point to the center of the fitted circle H and the radius of the circle R 350 The cumulative sum of the differences is taken as the fitting center error, and the fitting error is set to e: Where n represents the number of points in the point set; According to the above method, set the threshold t = 20, repeat the calculation t times, and store the fitting error in the set S e ={e t } t∈N , find the set S e The fitting circle center H corresponding to the minimum fitting error; S222, using the least squares nonlinear fitting Levenberg-Marquardt algorithm based on radius constraints to quickly iterate and calculate the optimal circle center, and transform the point set P R350 ={(x i ,z i )} i∈N , circle radius R 350 , the center of the circle H(x c ,z c ) is used as the input parameter of the algorithm iteration and the target residual function is defined as: Where M represents the number of fitting points. When F(x) takes the minimum value, the optimal center Q of the large arc is obtained. R350 ; S223, in the small arc point set P R20 In the example, we randomly select two points E and F, set the origin of the arc point cloud coordinate system as O, the midpoint of the chord EF as G, the center of the fitting circle as H, and the line segment GH as the perpendicular bisector of the chord EF. Calculate the vector and vector Model: in is the radius of the arc, and the vector is obtained based on the properties of similar triangles According to the known conditions, the center H is obtained: Assume that the coordinates of the fitting circle center H are (x c ,z c ), calculate the point set P R20 ={(x i ,z i )} i∈N The distance from each point to the center of the fitted circle H and the radius of the circle R 20 The cumulative sum of the differences is taken as the fitting center error, and the fitting error is set to e: Where n represents the number of points in the point set; According to the above method, set the threshold t = 20, repeat the calculation t times, and store the fitting error in the set S e ={e t } t∈N , find the set S e The fitting circle center H corresponding to the minimum fitting error; S224, using the least squares nonlinear fitting Levenberg-Marquardt algorithm based on radius constraints to quickly iterate and calculate the optimal circle center, and convert the point set P R20 ={(x i ,z i )} i∈N , circle radius R 20 , the center of the circle H(x c ,z c ) is used as the input parameter of the algorithm iteration and the target residual function is defined as: Where M represents the number of fitting points. When F(x) takes the minimum value, the optimal center Q of the small arc is obtained. R20 .

6. The rail wear detection method of full-section profile registration according to claim 1, characterized in that: The step S3 specifically comprises the following steps: S31, assuming that the optimal center coordinates of the large and small arcs are (Cx 350 ,Cz 350 ) and (Cx 20 ,Cz 20 ), the coordinates of the centers of the large and small arcs at the corresponding positions of the standard 50kg / m rail profile are (Cx B350 ,Cz B350 ) and (Cx B20 ,Cz B20 ), calculate the rotation angle θ, rotation matrix R, and translation matrix T for the registration of the original profile and the standard profile, and complete the registration of the inner profile: S32, assuming that the rotation and translation matrices of the medial profile registration are R I , T I , the inner profile calibration rotation matrix is ​​R Ibd , the rotation and translation matrices of the inner profile from the unified coordinate system and the standard profile coordinate system are obtained as R bz , T bz : R bz =R I *R Ibd T T bz =T I Assume that the calibration rotation and translation matrices of the outer contour are R Obd 、T Obd , before and after the registration, a certain point is (x i ,z i ) and (x i ',z i '), thus completing the outer contour registration: At this point, the full cross-section profile alignment of the inner and outer sides of the rail profile is completed.

7. The rail wear detection method based on full cross-section profile registration according to claim 1, characterized in that: The step S4 specifically comprises the following steps: S41. Measure the characteristic points on the inner and outer sides of the rail profile of the measured profile and the standard profile after registration, including the coordinates of the inner vertical wear points of the standard profile and the measured profile, respectively. bv ,z bv )、V'(x v ,z v ), the coordinates of the inner and outer rail jaw points of the standard profile are A in (x bie ,z bie ), A out (x boe ,z boe ), the coordinates of the inner and outer rail jaw points of the measured profile are A i ' n (x ie ,z ie ), A' out (x oe ,z oe ), the coordinates of the wear points on the inner and outer sides of the standard profile are H in (x bih ,z bih ), H out (x boh ,z boh ), the coordinates of the wear points on the inner and outer sides of the measured profile are H i ' n (x ih ,z ih ), H' out (x oh ,z oh ); S42. Calculate the vertical wear of the rail based on the characteristic points on the inner and outer sides of the rail profile. v 、Side wearω h and total wear ω a :

8. A rail wear detection device with full cross-section profile registration, characterized in that: include: The point cloud preprocessing and feature extraction module is used to cluster and filter the collected original track profile point cloud data based on the distance-based clustering and filtering algorithm, and then accurately extract the point cloud of the rail head and rail waist area based on the method of combining the longitudinal amplitude characteristics of the point cloud with the profile curvature characteristics; The optimal double center fitting module is used to segment the point cloud feature arcs in the inner rail waist area of ​​the rail profile to obtain large and small arcs, and to obtain the optimal double centers of the segmented large and small arcs through the nonlinear least squares algorithm based on radius constraints; The full-section profile registration module is used to calculate the inner side posture transformation parameters of the inner side profile of the rail according to the optimal dual-center positioning obtained by fitting, and calculate the outer side posture transformation parameters of the outer side profile of the rail by combining the inner side posture transformation parameters with the system calibration parameters, so as to realize the full-section profile registration of the measured profile and the standard profile; The rail wear value calculation module is used to calculate the rail wear value by combining the measured profile after profile registration with the characteristic points on the inner and outer sides of the rail profile of the standard profile.

9. 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 steps of the rail wear detection method with full-section profile registration as claimed in any one of claims 1 to 7 are implemented.

10. A storage medium, comprising a stored program, which controls a device where the storage medium is located to execute the steps of the rail wear detection method for full-section profile registration as claimed in any one of claims 1 to 7 when the program is executed.

Images