CN111932676A - Device and method for quickly measuring railway gauge - Google Patents
Device and method for quickly measuring railway gauge Download PDFInfo
- Publication number
- CN111932676A CN111932676A CN202010859950.5A CN202010859950A CN111932676A CN 111932676 A CN111932676 A CN 111932676A CN 202010859950 A CN202010859950 A CN 202010859950A CN 111932676 A CN111932676 A CN 111932676A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- template
- cloud
- sampling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 48
- 239000010959 steel Substances 0.000 claims abstract description 48
- 238000005259 measurement Methods 0.000 claims abstract description 31
- 238000005070 sampling Methods 0.000 claims description 35
- 239000011159 matrix material Substances 0.000 claims description 30
- 230000009466 transformation Effects 0.000 claims description 28
- 238000013500 data storage Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 10
- 238000013519 translation Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000000844 transformation Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 10
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000035874 Excoriation Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
Abstract
The invention discloses a device and a method for quickly measuring the railway track gauge, wherein an aircraft is used as a carrier, a measuring device is arranged at the bottom of the aircraft, the aircraft flies along a railway track in the space above the railway track, the measuring device collects original steel rail image data, pre-built steel rail BIM model point cloud is used as a template, and the inner side surface of the template point cloud is extracted and used as a reference surface on the basis of point cloud matching facing the template, so that the inner side surface of the actually measured steel rail point cloud is extracted, and the quick measurement of the track gauge is finally realized. The invention has the advantages of less occupied resources for measurement, short detection time, non-contact measurement and the like.
Description
Technical Field
The invention belongs to the technical field of optical detection, relates to a railway track detection technology, and particularly relates to a railway gauge rapid measuring device and a rapid measuring method thereof.
Background
Railway transportation has become the first choice mode of transportation of present masses' trip, and plays an important role in freight transportation, and the track is as the basis of railway operation, and the quick detection problem to railway environment infrastructure is non-negligible. The change of the track gauge of the track can cause various vibrations of the train, so that the acting force of the wheel track changes, the acting force is a control factor influencing the running safety and stability of the train in the aspect of the track and is also an important reason for the damage and the failure of track structural components, the track gauge detection is regularly carried out along with the increase of the running speed of the high-speed railway and the continuous expansion of the running scale, the track gauge state information is timely mastered, and the track gauge detection method becomes one of the most important contents in track detection projects.
Traditional contact gauge detection requires that railway maintainers place check out test set (like mechanical scale) on two strands of rail, and constantly impels forward along the railway trend, if measured gauge does not conform to the standard requirement, then the scale can be blocked, and this method not only need consume a large amount of manpower resources, and the measurement accuracy of scale is influenced by external conditions such as temperature, weather in addition, can't guarantee measuring accuracy nature.
The non-contact measurement is opposite to the contact measurement, namely, physical scanning equipment such as a camera, a sound wave instrument, a laser instrument and the like is carried on a track inspection vehicle (rail inspection vehicle for short) or a train body to capture the outline data of the steel rail so as to calculate the rail distance, and data acquisition equipment used by the measurement method is not in direct contact with the steel rail. In the method, the profile data of the steel rail is acquired by using the sensor and is used for calculating the gauge, the carrier needs to move on the rail so as to acquire the data, the rail resource needs to be occupied, the potential safety hazard of personnel exists, and the secondary abrasion of the steel rail is easily caused when the equipment carrier is contacted with the steel rail.
Unmanned aerial vehicle can be in contact with the rail completely, not occupy the data under the condition of track resource through carrying on camera equipment. In Vision-based rail track extraction and monitoring through the rail image of the railways, an unmanned aerial vehicle is used for acquiring a railway scene image, the rails are separated from the surrounding environment according to HSV colors, parallel steel rails are extracted by using Canny edge detection and nearest neighbor algorithm and are used for calculating the rail distance, and feasibility and efficiency of using an aircraft for rail detection are proved.
Disclosure of Invention
One of the purposes of the invention is to provide a railway track gauge measuring device which can realize non-contact rapid detection, does not occupy track resources and has low cost so as to meet the requirements of railway construction and maintenance departments on high-efficiency measurement of the rail gauge of a steel rail.
The technical scheme adopted by the invention for solving the technical problems is as follows: the utility model provides a quick measuring device of railway gauge, is including installing the three-dimensional camera in the aircraft bottom and the host system and the data acquisition module of being connected with the three-dimensional camera respectively, the data acquisition module on be connected with data storage module, the last data processing module that is connected with of data storage module, data storage module and data processing module be connected with host system respectively, host system control three-dimensional camera is shot the formation of image to the railway track, data acquisition module sends data storage module after sampling the image, host system control data processing module reads data from data storage module and analyzes, handles and give data storage module again and save, host system on still be connected with the power.
The second purpose of the invention is to provide a method for measuring the track gauge of a railway track, which comprises the following steps: step 1, data preprocessing
Step 11, establishing a standard BIM model of the steel rail according to the design requirements of the national standard on the steel rail based on the information of the rail head, the rail web, the rail bottom, the rail gauge and the inner side working surface below the rail head tread, and dispersing the standard BIM model into three-dimensional point cloud serving as template point cloud for matching;
step 12, constructing an actual measurement steel rail point cloud based on an actual measurement image acquired by an aircraft, and then segmenting the actual measurement steel rail point cloud from the scene point cloud to obtain the actual measurement point cloud;
step 21, performing point cloud primary matching based on SAC _ IA (sampling consistency initial registration) algorithm;
step 22, judging whether the distance error and the function reach the minimum, if not, repeating the step 21, and if so, stopping the current iteration;
step 23, performing accurate registration based on ICP (Iterative Closest Point) on the template Point cloud and the actually measured Point cloud obtained by the SAC _ IA algorithm;
step 24, judging whether the distance error is smaller than a threshold value or reaches the maximum iteration times;
step 25, if not, repeating the step 23, and if so, enabling the template point cloud and the actually-measured point cloud to be overlapped to the maximum extent to obtain the registered template point cloud and the actually-measured point cloud;
Step 31, taking an inner working surface in the template point cloud as a template reference surface, extracting inner points of the template point cloud, and calculating the template point cloud reference surface;
and step 32, extracting the inner side points of the actually measured point cloud, fitting to obtain an inner side working surface, calculating the track gauge of the actually measured point cloud on the basis of the inner side working surface, and outputting the track gauge.
In the step 21, Fast Point Feature Histograms (FPFH) of a source point cloud (namely, a template point cloud) and a target point cloud (namely, an actual measurement point cloud) are respectively calculated, feature descriptions of all points in a point set are obtained, then sampling point pairs of the template point cloud and the actual measurement point cloud are iteratively obtained, a transformation matrix from the source point cloud to the target point cloud is calculated according to the sampling point pairs, and finally, a distance error and the minimum transformation are selected as a final matching result.
Further, the specific steps are as follows:
step 211, calculating normal vectors and fast point feature histogram features of each point in the source point cloud P and the target point cloud Q respectively;
step 212, automatically selecting n sampling points P in the source point cloud Pk(k is 1,2, …, n) and guarantees any two sampling points pkiAnd pkjSatisfies the following formula:
wherein d isminIs a specified inter-dot distance threshold;
step 213, searching and sampling point p in the target point cloud Qk( k 1,2, …, n) corresponding sample points q with similar fast point feature histogram featureskAs one-to-one corresponding point of the source point cloud P in the target point cloud Q;
step 214, calculate the sampling point pkAnd corresponding sample point qkRigid body transformation matrix between, and sampling point pkPoint p obtained after rigid body transformationk' with corresponding sample point qkIs a difference of distance liThen there is li＝‖p′k-qk‖2Wherein p isk′(xk′,yk′,zk′)＝Rpk(xk,yk,zk) + T, where R represents a rotation matrix and T represents a translation matrix;
step 215, finally, an optimal set of transformations should be found, such that the distance error and the function
Is smallest, the matrix transformation at this point is considered as the final registration transformation matrix, wherein,
in the formula mlFor a given distance threshold,/iTransformed distance differences for the ith set of sample point pairs.
In the step 22 of the method for rapidly measuring the railway track gauge, corresponding relation point pairs are selected from a source point cloud and a target point cloud in an iterative manner, and an optimal rigid body transformation matrix between the point pairs is calculated until the convergence precision requirement, namely the distance error sum is minimum, is met. The difference is that when finding the corresponding point for each point in the source point cloud, the point closest to the point is selected as the corresponding point instead of the principle of random selection.
And the basic principle is to determine the corresponding point of the target point cloud in the nearest neighbor of the source point cloud, and then calculate the optimal rigid body transformation matrix between the point pairs until the convergence precision requirement is met, namely the distance error sum is minimum.
Further, the specific steps are as follows:
step 221, one sampling point set of the source point cloud is P ', and P ' is taken as each sampling point in P 'k(k is 1,2, …, N), finding its nearest corresponding point Q in the target point cloud Qk：
Step 222, calculating a rotation matrix R and a translation matrix T for the point set P' and the sampling point pair set of the target point cloud Q, so that the mean square error E (R, T) of the sampling point set is minimum:
step 223, using the current optimal transformation matrices R and T to the point set P' to obtain a new point set P ″:
p″k(x″k，y″k，z″k)＝Rp′k(x′ky′kz′k)+T
wherein, p ″)kRepresents p'kA transformed point;
step 224, calculate the mean distance error dAnd current iteration times I, if the convergence condition is met, the iteration is terminated, otherwise, the point set P 'is used for replacing the source point set P', and the step 221 is repeated to continue execution until the convergence condition is met:
I>Imax
in the formula, I is a specified tolerancemaxIs a specified maximum number of iterations.
3, the inner working surface of the template point cloud after the point cloud is primarily matched is used as a reference plane, a fitting plane most similar to the reference plane is searched and determined in the actually measured point cloud, and the fitting plane represents the inner side surface of the actually measured steel rail.
The method for rapidly measuring the railway gauge comprises the following steps of extracting the point cloud inner side points of the template point in the step 31:
the known template point cloud P is formed by two parallel steel rail model point clouds Pl，PrThe point cloud after registration with the actually measured point cloud is a source point cloud P', wherein Pl，PrRespectively correspond to P after registrationl′，Pr′；
For any two non-coincident vertexes P' in the source point cloud Pi′,pj', i ≠ j, and the distance between two points is:
d(p′i,p′j)＝||p′i-p′j||2；
for any point P in the source point cloud Pli', if the point is also PlOne point of' at Pr' A point p can be foundj' satisfies: min d (p'li,p′j)＝d0,p′j∈P′rIf the shortest distance from a certain point in the left steel rail to the right steel rail is equal to the standard gauge, the point is called as the inner side point of the left steel rail of the template point cloud;
similarly, for any point P in the source point cloud PriIf the point is PrOne point of' at Pl' A point p can be foundk' satisfies: min d (p)ri′,pk′)＝d0,pk′∈PlThe point is called as the inner point of the steel rail on the right side of the template point cloud.
In the method for measuring the track gauge of the railway track, the fitting step of the inner side surface of the template point cloud in the step 32 is as follows: an inner side point set of the template point cloud can be obtained through inner side point extraction:
Upl＝{pl1′,pl2′,pl3′,……,plm′}，Upr＝{pr1′,pr2′,pr3′,……,prn' }, two inner side planes of the template point cloud can be respectively obtained through least square plane fitting;
the method for extracting the inner side surface of the actually measured point cloud comprises the following steps: after the template point cloud and the actual measurement point cloud are registered, the shape and the position of the actual measurement point cloud are very similar to those of the template point cloud. And (3) taking the inner side surface of the known template point cloud as a reference surface, and searching and determining a fitting plane most similar to the reference surface in the actually measured point cloud to obtain the inner side surface of the actually measured point cloud.
The method for measuring the track gauge of the railway track comprises the following steps of:
collecting U of inner side points of left and right steel rails of actual measurement point cloud Q obtained after inner side surface extractionql，UqrRespectively is Sql，SqrIf the distance between a point q and a plane S is D (q, S) for any point q and a certain plane S in space, the track distance D of the measured point cloud is calculated according to the following formulaq：
Wherein n isl，nrRespectively represent a set UqlAnd UqrThe scale of (c); thus obtaining the track gauge of the track.
Compared with the prior art, the invention has the advantages that: the aircraft is used as a platform, and the rapid gauge measurement can be realized under the condition of not occupying railway resources by utilizing the advantages of high moving speed of the aircraft, large field range of the three-dimensional camera and rapid imaging; the rail gauge of the steel rail is measured in a non-contact mode, and the sensor and the steel rail cannot be abraded.
Drawings
FIG. 1 is a schematic diagram of relative positions of a steel rail template point cloud and an actually measured point cloud;
FIG. 2 is a partially enlarged view of the steel rail of FIG. 1;
FIG. 3 is a schematic block diagram of the structure of the measuring device of the present invention;
FIG. 4 is a general flow chart of the measurement method of the present invention;
FIG. 5 is a schematic view of an aircraft measuring gauge.
The figures are numbered: 1-rail head, 2-rail web, 3-rail base, 4-gauge, 5-inner working surface, 6-inner point of template point cloud, and 7-inner point of actual measurement point cloud.
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings: it should be understood that the preferred embodiments are illustrative of the invention only and are not limiting upon the scope of the invention.
Example 1
Referring to fig. 3, the device for rapidly measuring the rail gauge of the railway track comprises an aircraft provided with a measuring device, wherein the measuring device comprises a three-dimensional camera, a data acquisition module, a data storage module, a data processing module and a power supply.
The main control module is connected with the three-dimensional camera, the data acquisition module, the data storage module and the power supply, the power supply supplies power to the main control module, and the main control module controls the three-dimensional camera, the data acquisition module and the data storage module; the three-dimensional camera is connected with the data acquisition module; the data acquisition module is connected with the data storage module; the data storage module is connected with the data processing module.
The main control module controls the three-dimensional camera to photograph and image the railway track, the image is sampled by the data acquisition module and then sent to the data storage module, the main control module controls the data processing module to read the data from the data storage module for analysis and processing, and the processed data is sent to the data storage module for storage.
One of the typical application scenarios is as follows:
flight parameters of the aircraft: the flying speed is 5 m/s, the flying height is 24.8 m, the heading and side-to-side overlapping rates are 90 percent and 60 percent respectively, and the angle of the main route is 111 degrees.
Example 2
Referring to fig. 1 to 5, the method for rapidly measuring the track gauge of a railway track disclosed by the invention comprises the following steps:
and stage 1, preprocessing data.
Based on the information of the rail head 1, the rail web 2, the rail bottom 3, the rail gauge 4 and the inner side working surface 5 below the rail head tread, a standard BIM model of the rail is created according to the design requirements of the national standard on the rail, and is discretized into a three-dimensional point cloud serving as a template point cloud for matching. And meanwhile, constructing an actually measured scene point cloud aiming at an image set acquired by the aircraft, and then segmenting the actually measured steel rail point cloud from the scene to obtain the actually measured point cloud.
And 2, matching point clouds facing the template.
In the stage, point clouds of a standard BIM model are used as a template, and a template-oriented point cloud registration algorithm is utilized to perform registration on the template point cloud to the actually-measured steel rail point cloud in two steps, so that the two point clouds are overlapped to the maximum extent;
step 21, performing point cloud primary matching based on SAC _ IA (sampling consistency initial registration) algorithm;
step 22, judging whether the distance error and the function reach the minimum, if not, repeating the step 21, and if so, stopping the current iteration;
step 23, performing accurate registration based on ICP (Iterative Closest Point) on the template Point cloud and the actually measured Point cloud obtained by the SAC _ IA algorithm;
step 24, judging whether the distance error is smaller than a threshold value or reaches the maximum iteration times;
and 25, if not, repeating the step 23, and if so, enabling the template point cloud and the actually-measured point cloud to be overlapped to the maximum extent to obtain the registered template point cloud and the actually-measured point cloud.
The matching process at this stage is divided into two methods, the first method has its own set of iterative processes, which are described in step 21 and step 22: after the first method iteration is terminated, the result is used as input for the second method, and then a new, independent set of iterations is restarted using the second method, so that steps 23 and 24 are directed solely around the second method, i.e., ICP-based registration.
The method takes the point cloud of a standard BIM model as a template, and performs registration on the template point cloud to the actually measured steel rail point cloud in two steps by utilizing a template-oriented point cloud registration algorithm, so that the two point clouds are overlapped to the maximum extent.
The method comprises the steps of taking point cloud (hereinafter referred to as template point cloud) corresponding to a steel rail standard BIM model as source point cloud, taking actual measurement steel rail point cloud (hereinafter referred to as actual measurement point cloud) generated by an aircraft shooting image set as target point cloud, and realizing matching of the two point cloud and the target point cloud by adopting a template-oriented two-step registration algorithm. Wherein the two-step registration algorithm is: the method comprises the steps of firstly adopting an SAC _ IA algorithm to realize initial matching, ensuring that the template point cloud and the actually-measured point cloud have good relative initial positions, and then further realizing accurate matching based on an ICP algorithm.
The point cloud initial matching based on the SAC _ IA algorithm is to realize initial matching by adopting the SAC _ IA algorithm, so that the template point cloud and the actually measured point cloud are ensured to have better relative initial positions, and the basic idea is as follows: respectively calculating Fast Point Feature Histograms (FPFH) of the source point cloud and the target point cloud, obtaining feature descriptions of each point in the point set, then iteratively obtaining sampling point pairs of the template point cloud and the actually-measured point cloud, calculating a transformation matrix from the source point cloud to the target point cloud according to the sampling point pairs, and finally selecting the transformation with the minimum distance error as a final matching result. The method comprises the following specific steps:
(1) and respectively calculating the normal vector and the FPFH (field programmable gate flash) characteristic of each point in the source point cloud P and the target point cloud Q.
(2) Automatically selecting n sampling points P in source point cloud Pk(k is 1,2, …, n) and guarantees any two sampling points pkiAnd pkjSatisfies the following formula,
wherein d isminIs a specified inter-dot distance threshold.
(3) Searching and sampling point p in target point cloud Qk(k-1, 2, …, n) corresponding sample points q with similar FPFH characteristicskAnd the point points are used as the one-to-one corresponding points of the source point cloud P in the target point cloud Q.
(4) Calculating the corresponding point pk、qkA rigid body transformation matrix between, and pkPoint obtained after rigid body transformation (using p)k' represent) and point qkIs a difference of distance liThen, there are:
li＝‖p′k-qk‖2；
wherein p isk′(xk′,yk′,zk′)＝Rpk(xk,yk,zk) + T, R denotes a rotation matrix and T denotes a translation matrix.
(5) Finally, an optimal set of transformations should be found, such that the distance error sum function
Is the smallest (as shown in the above paragraph), the matrix transformation at this time is considered as the final registration transformation matrix (see this paragraph), wherein,
in the formula mlFor a given distance threshold,/iTransformed distance differences for the ith set of sample point pairs.
Point cloud fine matching based on an ICP algorithm: the transformation matrix obtained based on the SAC-IA algorithm is not accurate, and the matching effect of the template point cloud and the actually measured point cloud is difficult to ensure, so that the ICP algorithm is adopted to realize the accurate matching of the template point cloud and the actually measured point cloud. The basic idea of the algorithm is similar to the SAC _ IA algorithm, and the optimal rigid body transformation matrix between the point pairs is calculated by iteratively selecting corresponding relation point pairs from a source point cloud and a target point cloud until the convergence precision requirement, namely the distance error sum is minimum, is met. The difference is that when finding the corresponding point for each point in the source point cloud, the point closest to the point is selected as the corresponding point instead of the principle of random selection. And the basic principle is to determine the corresponding point of the target point cloud in the nearest neighbor of the source point cloud, and then calculate the optimal rigid body transformation matrix between the point pairs until the convergence precision requirement is met, namely the distance error sum is minimum. The method comprises the following specific steps:
(1) step 221, one sampling point set of the source point cloud is P ', and P ' is taken as each sampling point in P 'k(k ═ 1, 2.. N), its nearest corresponding point Q is sought in the target point cloud Qk：
(2) Calculating a rotation matrix R and a translation matrix T aiming at a sampling point pair set of a point set P' and a target point cloud Q, so that the mean square error E (R, T) of the sampling point set is minimum;
(3) using the current optimal transformation matrices R and T for the point set P' to obtain a new point set P ″:
p″k(x″k，y″k，z″k)＝Rp′k(x′k，y′k，z′k)+T
wherein, p ″)kRepresents p'kTransformed points.
(4) Calculating the mean value d of the distance errorsAnd current iteration times I, if the formula 3.8 or 3.9 is met, the iteration is terminated, otherwise, the point set P 'is used for replacing the point set P', and the step (3) is repeated to continue execution until the convergence condition is met:
I＞Imax；
in the formula, I is a specified tolerancemaxIs a specified maximum number of iterations.
And 3, calculating the track gauge based on the template reference surface.
And taking the inner side surface of the template point cloud as a reference surface, extracting the inner side points of the actually measured point cloud, fitting to obtain the inner side surface, and calculating the track gauge of the actually measured point cloud on the basis of the inner side surface.
Under the condition that the surface of the steel rail is not reconstructed, it is very difficult to accurately extract a vertex 16mm below a rail head tread from three-dimensional discrete point cloud according to a rail gauge definition (see fig. 1), and a fitting plane most similar to a reference plane is searched and determined in actually-measured point cloud by taking an inner working surface of template point cloud after point cloud registration as the reference plane, wherein the plane represents the inner side surface of the actually-measured steel rail. The inner point 6 of the template point cloud is on the inner working surface, and the inner point 7 of the actual measurement point cloud is on the inner working surface. Therefore, after the two point clouds are registered, the calculation of the track gauge is converted into a process of extracting a reference surface (namely, an inner working surface of the template point cloud) from the template point cloud, extracting an inner side surface based on the reference surface from the actual measurement point cloud, and calculating the track gauge from the extracted actual measurement working surface.
The method for extracting the template point cloud reference surface comprises the following steps:
(1) template point cloud interior point extraction
The known template point cloud P is formed by two parallel steel rail model point clouds Pl，PrThe point cloud after registration with the actually measured point cloud is P', wherein Pl，PrRespectively correspond to P after registrationl′，Pr′。
Definition 1: for any two non-coincident vertexes P in the point cloud Pi′,pj', i ≠ j, and the distance between two points is: d (p'i,p′j)＝||p′i-p′j||2。
Definition 2: for any point P in the point cloud Pli', if the point is also PlOne point of' at Pr' A point p can be foundj' satisfy：min d(p′li,p′j)＝d0,p′j∈P′r。
That is, the shortest distance from a certain point in the left steel rail to the right steel rail is equal to the standard gauge, and the point is called as the inner point of the left steel rail of the template point cloud.
Similarly, for any point P in the point cloud PriIf the point is PrOne point of' at Pl' A point p can be foundk' satisfies: min d (p)ri′,pk′)＝d0,pk′∈Pl' this point is said to be the inside point of the right rail of the template point cloud.
(2) Fitting of template point cloud inner side surface
By extracting the inner side points, the inner side point set of the template point cloud can be obtained
Upl＝{pl1′,pl2′,pl3′,……,plm′}，Upr＝{pr1′,pr2′,pr3′,……,prn' }, two inner side planes of the template point cloud can be respectively obtained through least square plane fitting.
The method for extracting the inner side surface of the actually measured point cloud comprises the following steps: after the template point cloud and the actual measurement point cloud are registered, the shape and the position of the actual measurement point cloud are very similar to those of the template point cloud. And (3) taking the inner side surface of the known template point cloud as a reference surface, and searching and determining a fitting plane most similar to the reference surface in the actually measured point cloud to obtain the inner side surface of the actually measured point cloud.
The method for calculating the track gauge of the actually measured point cloud comprises the following steps: after the inner side surface is extracted, the left and right steel rail inner side point set U of the obtained actual measurement point cloud Qql，UqrRespectively is Sql，Sqr. For any point q in space and a certain plane S, the distance between the point q and the plane S is D (q, S), and then the track distance D of the measured point cloud is calculated according to the following formulaq：
Wherein n isl，nrRespectively represent a set UqlAnd UqrThe scale of (a). Thus obtaining the track gauge of the track.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and it is apparent that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. The utility model provides a quick measuring device of railway gauge which characterized in that: including installing the three-dimensional camera in the aircraft bottom and main control module and the data acquisition module of being connected with the three-dimensional camera respectively, the data acquisition module on be connected with data storage module, be connected with data processing module on the data storage module, data storage module and data processing module be connected with main control module respectively, main control module control three-dimensional camera is to the formation of image of shooing of railway track, data acquisition module sends data storage module after sampling the image, main control module control data processing module reads data from data storage module and analyzes, handles and give data storage module again and save, main control module on still be connected with the power.
2. A method for rapid measurement of a measuring device according to claim 1, comprising the steps of:
step 1, data preprocessing
Step 11, establishing a standard BIM model of the steel rail based on information of a rail head (1), a rail web (2), a rail bottom (3), a rail gauge (4) and an inner side working surface (5) below a rail head tread, and dispersing the standard BIM model into three-dimensional point cloud serving as template point cloud for matching;
step 12, constructing an actual measurement steel rail point cloud based on the image acquired by the aircraft, and segmenting the actual measurement steel rail point cloud from the scene point cloud to obtain the actual measurement point cloud;
step 2, realizing point cloud matching facing to template
Step 21, carrying out point cloud primary matching based on SAC _ IA algorithm;
step 22, judging whether the distance error and the function reach the minimum, if not, repeating the step 21, and if so, stopping the current iteration;
step 23, carrying out ICP-based accurate registration on the template point cloud and the actually measured point cloud obtained by the SAC _ IA algorithm;
step 24, judging whether the distance error is smaller than a threshold value or reaches the maximum iteration times;
step 25, if not, repeating the step 23, and if so, enabling the template point cloud and the actually-measured point cloud to be overlapped to the maximum extent to obtain the registered template point cloud and the actually-measured point cloud;
step 3, calculating the track gauge based on the template reference surface
Step 31, taking an inner working surface (5) in the template point cloud as a template reference surface, extracting inner points of the template point cloud, and calculating the template point cloud reference surface;
and 32, extracting the inner side points of the actually measured point cloud, fitting to obtain an inner side working surface (5), calculating the track gauge of the actually measured point cloud, and outputting the track gauge.
3. The method as claimed in claim 2, wherein the step 21 is to calculate fast point feature histograms of the source point cloud and the target point cloud respectively, obtain feature descriptions of each point in the point set, then iteratively obtain sampling point pairs of the template point cloud and the measured point cloud, calculate a transformation matrix from the source point cloud to the target point cloud according to the sampling point pairs, and select a transformation with a minimum distance error as a final result of matching.
4. A method for quickly measuring the track gauge of a railway as claimed in claim 3, which includes the following steps
Step 211, calculating normal vectors and fast point feature histogram features of each point in the source point cloud P and the target point cloud Q respectively;
step 212, automatically selecting n sampling points P in the source point cloud Pk(k is 1,2, …, n) and guarantees any two sampling points pkiAnd pkjSatisfies the following formula:
wherein d isminIs a specified inter-dot distance threshold;
step 213, searching and sampling point p in the target point cloud Qk(k 1,2, …, n) corresponding sample points q with similar fast point feature histogram featureskAs one-to-one corresponding point of the source point cloud P in the target point cloud Q;
step 214, calculate the sampling point pkAnd corresponding sample point qkRigid body transformation matrix between, and sampling point pkPoint p obtained after rigid body transformationk' with corresponding sample point qkIs a difference of distance liThen there is li＝||p′k-qk||2Wherein p isk′(xk′，yk′，zk′)＝Rpk(xk，yk，zk) + T, where R represents a rotation matrix and T represents a translation matrix;
step 215, finally, an optimal set of transformations should be found, such that the distance error and the function
Is smallest, the matrix transformation at this point is considered as the final registration transformation matrix, wherein,
in the formula mlFor a given distance threshold,/iTransformed distance differences for the ith set of sample point pairs.
5. The method according to claim 4, wherein the step 22 is to calculate the optimal rigid transformation matrix between the point pairs by iteratively selecting the corresponding relationship point pairs from the source point cloud and the target point cloud until the convergence accuracy requirement, i.e. the distance error sum, is minimal.
6. A method for quickly measuring the track gauge of railway as claimed in claim 5, which includes such steps as
Step 221, one sampling point set of the source point cloud is P ', and P ' is taken as each sampling point in P 'k(k 1, 2.., N), finding its nearest corresponding point Q in the target point cloud Qk：
Step 222, calculating a rotation matrix R and a translation matrix T for the point set P' and the sampling point pair set of the target point cloud Q, so that the mean square error E (R, T) of the sampling point set is minimum:
step 223, using the current optimal transformation matrices R and T to the point set P' to obtain a new point set P ″:
p″k(x″k，y″k，z″k)＝Rp′k(x′k，y′k，z′k)+T
wherein, p ″)kRepresents p'kA transformed point;
step 224, calculate the mean distance error dAnd current iteration times I, if the convergence condition is met, the iteration is terminated, otherwise, the point set P 'is used for replacing the source point set P', and the step 221 is repeated to continue execution until the convergence condition is met:
I＞Imax
in the formula, I is a specified tolerancemaxIs a specified maximum number of iterations.
7. The method for rapidly measuring the railway track gauge according to claim 6, wherein in the step 3, an inner working surface (5) of the template point cloud after the point cloud is primarily matched is used as a reference plane, a fitting plane most similar to the reference plane is searched and determined in the actually measured point cloud, and the fitting plane represents the inner side surface of the actually measured steel rail.
8. The method for rapidly measuring the railway track gauge according to claim 7, wherein the step of extracting the inner points of the template point cloud in the step 31 comprises the following steps:
the known template point cloud P is formed by two parallel steel rail model point clouds Pl，PrThe point cloud after registration with the actually measured point cloud is a source point cloud P', wherein Pl，PrRespectively correspond to P after registrationl′，Pr′；
For any two non-coincident vertexes P' in the source point cloud Pi′，pj', i ≠ j, and the distance between two points is:
d(p′i，p′j)＝||p′i-p′j||2；
for any point P in the source point cloud Pli', if the point is also PlOne point of' at Pr' A point p can be foundj' satisfies: min d (p'li，p′j)＝d0，p′j∈P′rIf the shortest distance from a certain point in the left steel rail to the right steel rail is equal to the standard gauge, the point is called as the inner side point of the left steel rail of the template point cloud;
similarly, for any point P in the source point cloud PriIf the point is PrOne point of' at Pl' A point p can be foundk' satisfies: min d (p)ri′，pk′)＝d0，pk′∈PlThe point is called as the inner point of the steel rail on the right side of the template point cloud.
9. The method for rapidly measuring the railway track gauge according to claim 7, wherein the step 32 of fitting the inside of the template point cloud comprises the following steps: and extracting the inner side points to obtain an inner side point set of the template point cloud:
Upl＝{pl1′，pl2′，pl3′，……，plm′}，Upr＝{pr1′，pr2′，pr3′，……，prn' }, two inner side planes of the template point cloud can be respectively obtained through least square plane fitting.
10. The method as claimed in claim 7, wherein the step 32 of calculating the measured point cloud comprises:
collecting U of inner side points of left and right steel rails of actual measurement point cloud Q obtained after inner side surface extractionql，UqrRespectively is Sql，SqrIf the distance between a point q and a plane S is D (q, S) for any point q and a certain plane S in space, the track distance D of the measured point cloud is calculated according to the following formulaq：
Wherein n isl，nrRespectively represent a set UqlAnd UqrThe scale of (c); thus obtaining the track gauge of the track.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010859950.5A CN111932676B (en) | 2020-08-24 | 2020-08-24 | Railway track gauge rapid measurement device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010859950.5A CN111932676B (en) | 2020-08-24 | 2020-08-24 | Railway track gauge rapid measurement device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111932676A true CN111932676A (en) | 2020-11-13 |
| CN111932676B CN111932676B (en) | 2024-01-05 |
Family
ID=73305122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010859950.5A Active CN111932676B (en) | 2020-08-24 | 2020-08-24 | Railway track gauge rapid measurement device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111932676B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112682270A (en) * | 2020-12-21 | 2021-04-20 | 华能安阳能源有限责任公司 | Height measuring method for wind turbine generator |
| CN112862882A (en) * | 2021-01-28 | 2021-05-28 | 北京格灵深瞳信息技术股份有限公司 | Target distance measuring method, device, electronic apparatus and storage medium |
| CN113408074A (en) * | 2021-06-28 | 2021-09-17 | 吉林大学 | Wheel set tread parameter measuring method and device |
| CN114332212A (en) * | 2022-03-11 | 2022-04-12 | 中国铁路设计集团有限公司 | Track superelevation and front-back height detection method based on vehicle-mounted mobile laser point cloud |
| CN115824070A (en) * | 2023-02-23 | 2023-03-21 | 湖南睿图智能科技有限公司 | Method and device for measuring size of steel rail piece |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150371431A1 (en) * | 2013-01-29 | 2015-12-24 | Andrew Robert Korb | Methods for analyzing and compressing multiple images |
| CN108133458A (en) * | 2018-01-17 | 2018-06-08 | 视缘(上海)智能科技有限公司 | A kind of method for automatically split-jointing based on target object spatial point cloud feature |
| CN108657222A (en) * | 2018-05-03 | 2018-10-16 | 西南交通大学 | Railroad track gauge and horizontal parameters measurement method based on vehicle-mounted Lidar points cloud |
| CN110647798A (en) * | 2019-08-05 | 2020-01-03 | 中国铁路设计集团有限公司 | Automatic track center line detection method based on vehicle-mounted mobile laser point cloud |
| CN210478710U (en) * | 2019-07-16 | 2020-05-08 | 北京中云亿安数据科技研究院有限公司 | Railway infringement monitoring system |
| CN111429498A (en) * | 2020-03-26 | 2020-07-17 | 中国铁路设计集团有限公司 | Railway business line three-dimensional center line manufacturing method based on point cloud and image fusion technology |
-
2020
- 2020-08-24 CN CN202010859950.5A patent/CN111932676B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150371431A1 (en) * | 2013-01-29 | 2015-12-24 | Andrew Robert Korb | Methods for analyzing and compressing multiple images |
| CN108133458A (en) * | 2018-01-17 | 2018-06-08 | 视缘(上海)智能科技有限公司 | A kind of method for automatically split-jointing based on target object spatial point cloud feature |
| CN108657222A (en) * | 2018-05-03 | 2018-10-16 | 西南交通大学 | Railroad track gauge and horizontal parameters measurement method based on vehicle-mounted Lidar points cloud |
| CN210478710U (en) * | 2019-07-16 | 2020-05-08 | 北京中云亿安数据科技研究院有限公司 | Railway infringement monitoring system |
| CN110647798A (en) * | 2019-08-05 | 2020-01-03 | 中国铁路设计集团有限公司 | Automatic track center line detection method based on vehicle-mounted mobile laser point cloud |
| CN111429498A (en) * | 2020-03-26 | 2020-07-17 | 中国铁路设计集团有限公司 | Railway business line three-dimensional center line manufacturing method based on point cloud and image fusion technology |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112682270A (en) * | 2020-12-21 | 2021-04-20 | 华能安阳能源有限责任公司 | Height measuring method for wind turbine generator |
| CN112682270B (en) * | 2020-12-21 | 2023-01-31 | 华能安阳能源有限责任公司 | Height measuring method for wind turbine generator |
| CN112862882A (en) * | 2021-01-28 | 2021-05-28 | 北京格灵深瞳信息技术股份有限公司 | Target distance measuring method, device, electronic apparatus and storage medium |
| CN113408074A (en) * | 2021-06-28 | 2021-09-17 | 吉林大学 | Wheel set tread parameter measuring method and device |
| CN114332212A (en) * | 2022-03-11 | 2022-04-12 | 中国铁路设计集团有限公司 | Track superelevation and front-back height detection method based on vehicle-mounted mobile laser point cloud |
| CN115824070A (en) * | 2023-02-23 | 2023-03-21 | 湖南睿图智能科技有限公司 | Method and device for measuring size of steel rail piece |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111932676B (en) | 2024-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111932676B (en) | Railway track gauge rapid measurement device and method | |
| CN111080627B (en) | 2D +3D large airplane appearance defect detection and analysis method based on deep learning | |
| CN109900712B (en) | Wind power blade mechanical fault on-line detection system and method based on camera measurement | |
| WO2016015546A1 (en) | System and method for aircraft docking guidance and aircraft type identification | |
| US11915408B2 (en) | Methods of artificial intelligence-assisted infrastructure assessment using mixed reality systems | |
| US10942130B2 (en) | Damage information processing device and damage information processing method | |
| WO2019222255A1 (en) | Computer aided inspection system and methods | |
| CN110866969A (en) | Engine blade reconstruction method based on neural network and point cloud registration | |
| Xie et al. | RRCNet: Rivet region classification network for rivet flush measurement based on 3-D point cloud | |
| CN110619678A (en) | Electric power tower inspection track generation method based on point cloud matching | |
| Minghui et al. | Deep learning enabled localization for UAV autolanding | |
| CN109927935A (en) | A kind of airframe upper surface inspection method that unmanned plane oblique photograph camera combines | |
| CN110942077B (en) | Feature line extraction method based on weight local change degree and L1 median optimization | |
| US11875524B2 (en) | Unmanned aerial vehicle platform based vision measurement method for static rigid object | |
| Liu et al. | A UAV-based aircraft surface defect inspection system via external constraints and deep learning | |
| Oskouie et al. | A data quality-driven framework for asset condition assessment using LiDAR and image data | |
| Yang et al. | Discretization–filtering–reconstruction: railway detection in images for navigation of inspection UAV | |
| Wu et al. | Autonomous UAV landing system based on visual navigation | |
| CN103853893A (en) | Parameter model matching method for detecting airplane attitude from point cloud data | |
| CN116202487A (en) | Real-time target attitude measurement method based on three-dimensional modeling | |
| Mahmoud et al. | Low-cost framework for 3D reconstruction and track detection of the railway network using video data | |
| CN112396604B (en) | Multi-view-angle-based aircraft skin defect detection method | |
| CN115131688A (en) | Unmanned aerial vehicle shooting point extraction method for inspection component | |
| Zhong et al. | Point cloud classification for detecting roadside safety attributes and distances | |
| CN112581521B (en) | Method for extracting central line of magnetic suspension track |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |