CN111932676B - Railway track gauge rapid measurement device and method - Google Patents
Railway track gauge rapid measurement device and method Download PDFInfo
- Publication number
- CN111932676B CN111932676B CN202010859950.5A CN202010859950A CN111932676B CN 111932676 B CN111932676 B CN 111932676B CN 202010859950 A CN202010859950 A CN 202010859950A CN 111932676 B CN111932676 B CN 111932676B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- template
- cloud
- track gauge
- 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.)
- Active
Links
- 238000005259 measurement Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims description 30
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 40
- 239000010959 steel Substances 0.000 claims abstract description 40
- 238000000691 measurement method Methods 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims description 39
- 239000011159 matrix material Substances 0.000 claims description 31
- 230000009466 transformation Effects 0.000 claims description 31
- 238000013500 data storage Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 8
- 238000013519 translation Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000000844 transformation Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 12
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 208000035874 Excoriation Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical 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
- 230000006872 improvement Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 235000021184 main course Nutrition 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a railway track gauge rapid measurement device and a railway track gauge rapid measurement method, wherein an aircraft is used as a carrier, a measurement device is arranged at the bottom of the aircraft, the aircraft flies in a space above a railway track and along the track, the measurement device collects original steel rail image data, a pre-built steel rail BIM model point cloud is used as a template, the inner side surface of the template point cloud is extracted and is used as a reference surface on the basis of point cloud matching facing the template, and therefore the inner side surface of the actually measured steel rail point cloud is extracted, and finally the rapid measurement of the track gauge is realized. The invention has the advantages of less resources occupation, short detection time, non-contact measurement and the like.
Description
Technical Field
The invention belongs to the technical field of optical detection, relates to railway track detection technology, and in particular relates to a railway track gauge rapid measurement device and a rapid measurement method thereof.
Background
Railway transportation is the first choice traffic mode for the current mass travel, plays an important role in cargo transportation, and the rail is used as the basis of railway operation and cannot be ignored for the rapid detection of railway environment infrastructure. The change of the track gauge can cause various vibrations of the train, so that the wheel track acting force is changed, the wheel track acting force is a control factor affecting the running safety and stability of the train in the aspect of the track, and is also an important cause of damage and failure of track structural components, the track gauge detection is carried out regularly along with the improvement of the running speed of a high-speed railway and the continuous expansion of the running scale, and the track gauge state information is grasped in time, so that the wheel track acting force is one of the most important contents in a track detection project.
Traditional contact gauge detection requires that railway maintenance personnel place detection equipment (such as a mechanical scale) on two rails, continuously push forward along the railway trend, and if the measured gauge does not meet the standard requirement, the scale can be clamped, so that the method not only needs to consume a large amount of manpower resources, but also the measurement accuracy of the scale is influenced by external conditions such as temperature, climate and the like, and the measurement accuracy cannot be guaranteed.
In contrast to contact measurement, non-contact measurement is performed, that is, physical scanning equipment, such as a camera, a sonic meter, a laser meter, etc., is mounted on a rail detection vehicle (rail detection vehicle for short) or a train body, so that rail profile data is captured to calculate the track gauge, and the data acquisition equipment used in such measurement methods does not directly contact the rail. The method utilizes the sensor to acquire the profile data of the steel rail and is used for calculating the track gauge, the carrier needs to move on the track to acquire the data, the carrier needs to occupy the track resource, the potential safety hazard of personnel exists, and the contact of the equipment carrier and the steel rail easily causes secondary abrasion of the steel rail.
The unmanned aerial vehicle can collect data under the condition of not contacting the steel rail and occupying no track resource by carrying the camera equipment. In "Vision based rail track extraction and monitoring through drone imagery", arun Kumar Singh et al acquires a railway scene image by using an unmanned aerial vehicle, separates a track from the surrounding environment according to HSV colors, extracts parallel steel rails by using Canny edge detection and nearest neighbor algorithm and is used for calculating the track gauge, and proves the feasibility and high efficiency of track detection by using an aircraft.
Disclosure of Invention
The invention aims 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 track gauge of a steel rail.
The technical scheme adopted for solving the technical problems is as follows: the utility model provides a railway track gauge rapid measurement device, includes the three-dimensional camera of installing in the aircraft bottom and respectively with three-dimensional camera connected main control module and data acquisition module, 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 imaged the railway track, data acquisition module gives data storage module after to the image sampling, main control module control data processing module reads data from data storage module and analyzes, handle and give data storage module again and preserve, main control module on still be connected with the power.
The second purpose of the invention is to provide a railway track gauge measuring method, which comprises the following steps: step 1, data preprocessing
Step 11, creating a standard BIM model of the steel rail according to the design requirements of national standards on the steel rail based on the information of the rail head, rail web, rail bottom, rail gauge and inner side working surface below the rail head tread of the steel rail, discretizing into a three-dimensional point cloud serving as a 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 dividing the actual measurement steel rail point cloud from a scene point cloud to obtain an actual measurement point cloud;
step 2, matching point clouds facing templates: the point cloud of the standard BIM model is used as a template, and the point cloud of the template is registered to the actually measured steel rail point cloud in two steps by utilizing a point cloud registration algorithm facing the template, so that the two point clouds are overlapped to the greatest extent;
step 21, performing point cloud initial 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 step 21, and if so, stopping the current iteration;
step 23, carrying out accurate registration based on ICP (Iterative Closest Point ) by using a template point cloud and a real point cloud obtained by a SAC_IA algorithm;
step 24, judging whether the distance error is smaller than a threshold value or the maximum iteration number is reached;
step 25, if not, repeating the step 23, and if so, enabling the template point cloud and the real point cloud to coincide to the greatest extent, so as to obtain the registered template point cloud and real point cloud;
step 3, track gauge calculation based on template reference surface
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 points of the real-point cloud, fitting to obtain an inner working surface, calculating the real-point cloud track gauge based on the inner points, and outputting the track gauge.
In the method, in step 21, fast Point Feature Histograms (FPFH) of a source point cloud (i.e., a template point cloud) and a target point cloud (i.e., a real point cloud) are calculated respectively, feature descriptions of each point in a point set are obtained, then sampling point pairs of the template point cloud and the real point cloud are obtained iteratively, 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 minimum transformation are selected as final matching results.
Further, the specific steps are as follows:
step 211, calculating normal vector and fast point characteristic 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 from the source point cloud P k (k=1, 2, …, n) and ensures any two sampling points p ki And p kj Satisfies the following formula:
wherein d min A threshold value for a specified inter-point distance;
step 213, finding and sampling point p in target point cloud Q k (k=1, 2, …, n) corresponding sample points q having similar fast point feature histogram features k As a one-to-one correspondence point of the source point cloud P in the target point cloud Q;
step 214, calculating the sampling point p k And corresponding sampling point q k Rigid body transformation matrix between and sampling point p k The point p obtained after rigid transformation k ' and corresponding sample point q k Distance difference l of (2) i Then there is l i ＝‖p′ k -q k ‖ 2 Wherein p is k ′(x k ′,y k ′,z k ′)＝Rp k (x k ,y k ,z k ) +T, where R represents a rotation matrix and T represents a translation matrix;
step 215, eventually a set of optimal transformations should be found, such that the distance error and the functionThe matrix transformation at this time is considered to be the final registration transformation matrix, where,
m is in l For a given distance threshold, l i And the distance difference after the transformation is the i-th group of sampling point pairs.
In the method for rapidly measuring the railway track gauge, in step 22, corresponding relation point pairs are selected from source point clouds and target point clouds in an iterative mode, and an optimal rigid body transformation matrix between the point pairs is calculated until convergence accuracy requirements, namely distance errors and minimum distances are met. The difference is that when searching for a corresponding point for each point in the source point cloud, the point closest to the point is selected as the corresponding point instead of according to the principle of random selection.
And the basic principle is that corresponding points of the target point cloud are determined in the nearest neighbor of the source point cloud, and then an optimal rigid body transformation matrix between the point pairs is calculated until the convergence accuracy requirement, namely the sum of distance errors is minimum, is met.
Further, the specific steps are as follows:
step 221, a sampling point set of the source point cloud is P ', for each sampling point P ' in P ' k (k=1, 2, …, N), finding its nearest counterpart point Q in the target point cloud Q k ：
Step 222, for the set of sampling point pairs of the point set P' and the target point cloud Q, calculating a rotation matrix R and a translation matrix T, so as to minimize a mean square error E (R, T) of the sampling point set:
step 223, using the current optimal transformation matrices R and T for the point set P', obtaining a new point set p″:
p″ k (x″ k ，y″ k ，z″ k )＝Rp′ k (x′ k y′ k z′ k )+T
wherein p' k Represents p' k Transformed points;
step 224, calculate the distance error mean d ε And the current iteration number 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 until the convergence condition is met:
I>I max
wherein ε is a specified allowable error, I max For a specified maximum number of iterations.
The step 3 of the rapid measurement method of the railway track gauge is to take the inner side working surface of the template point cloud after initial matching of the point cloud as a reference plane, search and determine a fitting plane most similar to the reference plane in the real-time point cloud, and represent the inner side surface of the actually measured steel rail by the fitting plane.
The method for rapidly measuring the railway track gauge comprises the following steps of:
the known template point cloud P is formed by two parallel steel rail model point clouds P l ，P r The point cloud registered with the real point cloud is a source point cloud P', wherein P is the point cloud P l ，P r Registered and respectively correspond to P l ′，P r ′；
For any two non-coincident vertexes P in the source point cloud P i ′,p j ' i not equal to j, 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 P li ' if this point is also P l ' one point in P r One point p can be found in j ' satisfy: min d (p' li ,p′ j )＝d 0 ,p′ j ∈P′ r Namely, the shortest distance from a certain point in the left steel rail to the right steel rail is equal to the standard track gauge, and 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 P ri ' if the point is P r ' one point in P l One point p can be found in k ' satisfy: min d (p) ri ′,p k ′)＝d 0 ,p k ′∈P l The point is the inboard point of the right rail of the template point cloud.
The fitting step of the inner side surface of the template point cloud in the step 32 of the railway track gauge measuring method comprises the following steps: the inboard point set of the template point cloud can be obtained through inboard point extraction:
U pl ＝{p l1 ′,p l2 ′,p l3 ′,……,p lm ′}，U pr ＝{p r1 ′,p r2 ′,p r3 ′,……,p rn ' two inner side planes of the template point cloud can be respectively obtained through least square plane fitting;
the extraction method of the cloud inner side surface of the actual measurement point comprises the following steps: after the template point cloud and the real point cloud are registered, the shape and the position of the real point cloud are very similar to those of the template point cloud. The inner side surface of the template point cloud is known and is used as a reference surface, and the fitting plane which is the most similar to the reference surface is searched and determined in the real point cloud, namely the inner side surface of the real point cloud.
The method for measuring the track gauge of the railway track comprises the following steps of:
left and right rail inner side point set U of real side point cloud Q obtained through inner side face extraction ql ，U qr Is a flat part of (2)The surface fitting equation is S ql ，S qr For any point q and a certain plane S in the space, if the distance from the point q to the plane S is D (q, S), the track gauge D of the real point cloud is calculated according to the following formula q ：
Wherein n is l ，n r Respectively represent the collection U ql And U qr Scale of (2); the track gauge of the track is thus obtained.
Compared with the prior art, the invention has the advantages that: the aircraft is adopted as a platform, and the advantages of high moving speed, large field of view of the three-dimensional camera and rapid imaging are utilized, so that the rapid track gauge measurement can be realized under the condition of not occupying railway resources; the rail gauge of the steel rail is measured in a non-contact mode, so that the sensor and the steel rail cannot be worn.
Drawings
FIG. 1 is a schematic diagram of the relative positions of a rail template point cloud and a real-time point cloud;
FIG. 2 is an enlarged view of a portion of the rail of FIG. 1;
FIG. 3 is a schematic block diagram 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 illustration of an aircraft measurement gauge.
The reference numerals are as follows: 1-rail head, 2-rail waist, 3-rail bottom, 4-rail gauge, 5-inner working face, 6-inner point of template point cloud and 7-inner point of real point cloud.
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the attached drawings: it should be understood that the preferred embodiments are presented by way of illustration only and not by way of limitation.
Example 1
Referring to fig. 3, the rapid measuring device for the track 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 enable the three-dimensional camera to photograph and image the railway track, the images are sampled by the data acquisition module and then sent to the data storage module, the main control module controls the data processing module to enable the main control module to read data from the data storage module for analysis and processing, and the processed data are sent to the data storage module for storage.
One of the typical application scenarios is as follows:
aircraft flight parameters: the flight speed was 5 m/s, the flight height was 24.8 m, the heading and side lap rates were 90% and 60%, respectively, and the main course angle was 111 °.
Example 2
Referring to fig. 1 to 5, the method for rapidly measuring the track gauge of the railway disclosed by the invention comprises the following measuring steps:
stage 1, data preprocessing.
Based on the information of the rail head 1, the rail web 2, the rail bottom 3, the rail gauge 4 and the inner 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 discretized into a three-dimensional point cloud serving as a template point cloud for matching. Meanwhile, aiming at an image set acquired by the aircraft, an actual measurement scene point cloud is constructed, and then the actual measurement rail point cloud is segmented from the scene to obtain an actual measurement point cloud.
And 2, matching the point cloud facing the template.
In the stage, the point cloud of the standard BIM model is used as a template, and the point cloud registration algorithm facing the template is utilized to register the point cloud of the template to the actually measured steel rail point cloud in two steps, so that the two point clouds are overlapped to the greatest extent;
step 21, performing point cloud initial 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 step 21, and if so, stopping the current iteration;
step 23, carrying out accurate registration based on ICP (Iterative Closest Point ) by using a template point cloud and a real point cloud obtained by a SAC_IA algorithm;
step 24, judging whether the distance error is smaller than a threshold value or the maximum iteration number is reached;
and step 25, if not, repeating the step 23, and if so, enabling the template point cloud and the real point cloud to coincide to the greatest extent, so as to obtain the registered template point cloud and real point cloud.
The matching process in this stage is divided into two methods, the first method has its own set of iterative processes, namely, the one described in step 21 and step 22: after the first method iteration is terminated, the result is taken as input to the second method, and then a new, independent set of iterations is restarted using the second method, so steps 23 and 24 are around the second method alone, i.e. ICP-based registration.
According to the invention, the point cloud of the standard BIM model is used as a template, and the point cloud registration algorithm facing the template is utilized to register the point cloud of the template to the actually measured steel rail point cloud in two steps, so that the two point clouds are overlapped to the greatest extent.
The method is characterized in that a point cloud corresponding to a steel rail standard BIM model (hereinafter referred to as a template point cloud) is used as a source point cloud, an actual measurement steel rail point cloud (hereinafter referred to as a real measurement point cloud) generated by an aircraft shooting image set is used as a target point cloud, and a two-step registration algorithm facing the template is adopted to realize the matching of the two. The two-step registration algorithm is as follows: firstly, an SAC_IA algorithm is adopted to realize initial matching, a template point cloud and a real point cloud are guaranteed to have good relative initial positions, and then, accurate matching is further realized 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 actual point cloud are ensured to have better relative initial positions, and the basic idea is that: and respectively calculating Fast Point Feature Histograms (FPFH) of the source point cloud and the target point cloud, acquiring feature description of each point in the point set, then iteratively acquiring sampling point pairs of the template point cloud and the real 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 a distance error and minimum transformation as a final matching result. The method comprises the following specific steps:
(1) And respectively calculating the normal vector and FPFH 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 P k (k=1, 2, …, n) and ensures any two sampling points p ki And p kj The following formula is satisfied,
wherein d min Is a specified inter-dot distance threshold.
(3) Searching for a sampling point p in a target point cloud Q k (k=1, 2, …, n) corresponding sample points q with similar FPFH characteristics k As a one-to-one correspondence point of the source point cloud P in the target point cloud Q.
(4) Calculating the corresponding point p k 、q k Rigid body transformation matrix between and p k The points obtained after rigid transformation (p k ' representation) and point q k Distance difference l of (2) i The following steps are:
l i ＝‖p′ k -q k ‖ 2 ；
wherein p is k ′(x k ′,y k ′,z k ′)＝Rp k (x k ,y k ,z k ) +T, R represents the rotation matrix and T represents the translation matrix.
(5) Eventually a set of optimal transformations should be found such that the distance error and the functionThe value of (a) is the smallest (as shown in the previous equation), and the matrix is transformed at this timeConsider the final registration transformation matrix (see this paragraph formula), where,
m is in l For a given distance threshold, l i And the distance difference after the transformation is the i-th group of sampling point pairs.
Point cloud exact matching based on 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 actual point cloud is difficult to ensure, so that the accurate matching of the template point cloud and the actual point cloud is realized by adopting the ICP algorithm. 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 the corresponding relation point pairs in the source point cloud and the target point cloud until the convergence accuracy requirement, namely the distance error sum is minimum, is met. The difference is that when searching for a corresponding point for each point in the source point cloud, the point closest to the point is selected as the corresponding point instead of according to the principle of random selection. And the basic principle is that corresponding points of the target point cloud are determined in the nearest neighbor of the source point cloud, and then an optimal rigid body transformation matrix between the point pairs is calculated until the convergence accuracy requirement, namely the sum of distance errors is minimum, is met. The method comprises the following specific steps:
(1) Step 221, a sampling point set of the source point cloud is P ', for each sampling point P ' in P ' k (k=1, 2,..n), finding its nearest corresponding point Q in the target point cloud Q k ：
(2) Aiming at a sampling point pair set of the point set P' and the target point cloud Q, calculating a rotation matrix R and a translation matrix T, so that the mean square error E (R, T) of the sampling point set is minimum;
(3) The current optimal transformation matrix R and T is used for the point set P ', so that a new point set P':
p″ k (x″ k ，y″ k ，z″ k )＝Rp′ k (x′ k ，y′ k ，z′ k )+T
wherein p' k Represents p' k Transformed points.
(4) Calculating a distance error mean value d ε And the current iteration number I, if the formula 3.8 or 3.9 is satisfied, the iteration is terminated, otherwise, the point set P 'is replaced by the point set P', and the step (3) is repeated until the convergence condition is satisfied:
I＞I max ；
wherein ε is a specified allowable error, I max For a specified maximum number of iterations.
Stage 3, track gauge calculation 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 real point cloud, fitting to obtain the inner side surface, and calculating the track gauge of the real point cloud based on the inner side surface.
In the case of not reconstructing the surface of the rail, it is very difficult to accurately extract the vertex 16mm below the tread of the rail head from the three-dimensional discrete point cloud according to the track gauge definition (see fig. 1), and the inner working surface of the template point cloud after the point cloud registration is used as a reference plane, and the fitting plane most similar to the reference plane is searched and determined in the real-time point cloud, wherein the plane represents the inner side surface of the actually measured rail. The inner side point 6 of the template point cloud is on the inner side working surface, and the inner side point 7 of the real point cloud is on the inner side working surface. Therefore, after the two point clouds are registered, the calculation of the gauge is converted into a process of extracting a reference surface (i.e., an inner working surface of the template point cloud) in the template point cloud, extracting an inner side surface based on the reference surface in the real point cloud, and calculating the gauge from the extracted actually measured working surface.
The template point cloud reference surface extraction method comprises the following steps:
(1) Template point cloud inside point extraction
The known template point cloud P consists of two parallel steel rail model point clouds P l ，P r The point cloud after being registered with the actual point cloud is P', wherein P is the point cloud l ，P r Registered and respectively correspond to P l ′，P r ′。
Definition 1: for any two non-coincident vertexes P in the point cloud P i ′,p j ' i not equal to j, 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 P li ' if this point is also P l ' one point in P r One point p can be found in j ' satisfy: min d (p' li ,p′ j )＝d 0 ,p′ j ∈P′ r 。
Namely, 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 the inner side point of the left steel rail of the template point cloud.
Similarly, for any point P in the point cloud P ri ' if the point is P r ' one point in P l One point p can be found in k ' satisfy: min d (p) ri ′,p k ′)＝d 0 ,p k ′∈P l The point is called as the inner side point of the right steel rail of the template point cloud.
(2) Template point cloud medial surface fitting
The inboard point set of the template point cloud can be obtained through inboard point extraction
U pl ＝{p l1 ′,p l2 ′,p l3 ′,……,p lm ′}，U pr ＝{p r1 ′,p r2 ′,p r3 ′,……,p rn ' two inner side planes of the template point cloud can be obtained respectively through least square plane fitting.
The method for extracting the cloud inner side surface of the real measurement point comprises the following steps: after the template point cloud and the real point cloud are registered, the shape and the position of the real point cloud are very similar to those of the template point cloud. The inner side surface of the template point cloud is known and is used as a reference surface, and the fitting plane which is the most similar to the reference surface is searched and determined in the real point cloud, namely the inner side surface of the real point cloud.
The method for calculating the cloud track gauge of the actual measurement point comprises the following steps: after the inner side face is extracted, a left steel rail inner side point set U of the real-time point cloud Q is obtained ql ，U qr Plane fitting equations of S ql ，S qr . For any point q and a certain plane S in the space, if the distance from the point q to the plane S is D (q, S), the track gauge D of the real point cloud is calculated according to the following formula q ：
Wherein n is l ，n r Respectively represent the collection U ql And U qr Scale of (c). The track gauge of the track is thus obtained.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, and it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit and scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (5)
1. A rapid measurement method of a rapid measurement device for railway track gauge is characterized in that,
based on a measuring device comprising a three-dimensional camera arranged at the bottom of an aircraft, a main control module and a data acquisition module, wherein the main control module and the data acquisition module are respectively connected with the three-dimensional camera, the data acquisition module is connected with a data storage module, the data storage module is connected with a data processing module, the data storage module and the data processing module are respectively connected with the main control module, the main control module controls the three-dimensional camera to photograph and image a railway track, the data acquisition module samples an image and then sends the image to the data storage module, and the main control module controls the data processing module to read data from the data storage module, analyze and process the data and then send the data to the data storage module for storage;
the method comprises the following steps:
step 1, data preprocessing
Step 11, creating a standard BIM model of the steel rail 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, and discretizing the model into a three-dimensional point cloud serving as a template point cloud for matching;
step 12, constructing an actual measurement steel rail point cloud based on an image acquired by an aircraft, and dividing the actual measurement steel rail point cloud from a scene point cloud to obtain an actual measurement point cloud;
step 2, realizing point cloud matching for templates
Step 21, performing point cloud initial matching based on SAC_IA algorithm: respectively calculating fast point characteristic histograms of a source point cloud and a target point cloud, acquiring characteristic description of each point in a point set, then iteratively acquiring sampling point pairs of a template point cloud and a real point cloud, calculating a transformation matrix from the source point cloud to the target point cloud according to the sampling point pairs, and selecting a distance error and minimum transformation as a final matching result;
step 211, calculating normal vector and fast point characteristic 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 from the source point cloud P k (k=1, 2, …, n) and ensures any two sampling points p ki And p kj Satisfies the following formula:
wherein d is min A threshold value for a specified inter-point distance;
step 213, finding and sampling point p in target point cloud Q k (k=1, 2, …, n) corresponding sample points q having similar fast point feature histogram features k As a one-to-one correspondence point of the source point cloud P in the target point cloud Q;
step 214, calculating the sampling point p k And corresponding sampling point q k Rigid body transformation matrix between and sampling point p k The point p obtained after rigid transformation k ' and corresponding sample point q k Distance difference l of (2) i Then there is l i ＝||p′ k -q k || 2 Wherein p is k ′(x k ′,y k ′,z k ′)＝Rp k (x k ,y k ,z k ) +T, where R represents a rotation matrix and T represents a translation matrix;
step 215, eventually a set of optimal transformations should be found, such that the distance error and the functionThe matrix transformation at this time is considered to be the final registration transformation matrix, where,
m is in l For a given distance threshold, l i The distance difference after the i-th group sampling point pair is transformed;
step 22, judging whether the distance error and the function reach the minimum, if not, repeating step 21, and if so, stopping the current iteration;
selecting corresponding relation point pairs from the source point cloud and the target point cloud iteratively, and calculating an optimal rigid body transformation matrix between the point pairs until convergence accuracy requirements, namely distance errors and minimum, are met;
step 221, a sampling point set of the source point cloud is P ', for each sampling point P ' in P ' k (k=1, 2, …, N), finding its nearest counterpart point Q in the target point cloud Q k ：
Step 222, for the set of sampling point pairs of the point set P' and the target point cloud Q, calculating a rotation matrix R and a translation matrix T, so as to minimize a mean square error E (R, T) of the sampling point set:
step 223, using the current optimal transformation matrices R and T for the point set P', obtaining a new point set p″:
p" k (x" k ,y" k ,z" k )＝Rp' k (x′ k ,y' k ，z′ k )+T
wherein p' k Represents p' k Transformed points;
step 224, calculate the distance error mean d ε And the current iteration number 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 until the convergence condition is met:
I＞I max
wherein ε is a specified allowable error, I max For a specified maximum number of iterations;
step 23, carrying out ICP-based accurate registration on the template point cloud and the actual point cloud obtained by the SAC_IA algorithm;
step 24, judging whether the distance error is smaller than a threshold value or the maximum iteration number is reached;
step 25, if not, repeating the step 23, and if so, enabling the template point cloud and the real point cloud to coincide to the greatest extent, so as to obtain the registered template point cloud and real point cloud;
step 3, track gauge calculation based on 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 step 32, extracting the inner points of the real-point cloud, fitting to obtain an inner working surface (5), calculating the track gauge of the real-point cloud, and then outputting the track gauge.
2. The method for rapidly measuring the railway track gauge according to claim 1, wherein the step 3 is to use an inner working surface (5) of a template point cloud after initial matching of the point cloud as a reference plane, search and determine a fitting plane most similar to the reference plane in a real-time point cloud, and represent the inner side surface of a measured steel rail by the fitting plane.
3. The method for rapidly measuring railway track gauge according to claim 2, wherein the step of extracting the inner points of the template point cloud in the step 31 is as follows:
the known template point cloud P is formed by two parallel steel rail model point clouds P l ，P r The point cloud registered with the real point cloud is a source point cloud P', wherein P is the point cloud P l ，P r Registered and respectively correspond to P l ′，P r ′；
For any two non-coincident vertexes P in the source point cloud P i ′,p j ' i not equal to j, 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 P li ' if this point is also P l ' one point in P r One point p can be found in j ' satisfy: min d (p' li ，p′ j )＝d 0 ，p′ j ∈P r ' namely, the shortest distance from a certain point in the left steel rail to the right steel rail is equal to the standard track gauge, and 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 P ri ' if the point is P r ' one point in P l One point p can be found in k ' satisfy: mind (p) ri ′,p k ′)＝d 0 ,p k ′∈P l The point is the inboard point of the right rail of the template point cloud.
4. The method for rapidly measuring railway track gauge according to claim 2, wherein the step of fitting the inner side of the template point cloud in the step 32 is as follows: extracting an inner side point set of the template point cloud through the inner side points: u (U) pl ＝{p l1 ′，p l2 ′，p l3 ′，……，p lm ′}，U pr ＝{p r1 ′，p r2 ′，p r3 ′，……，p rn ' two inner side planes of the template point cloud can be obtained respectively through least square plane fitting.
5. The method for rapidly measuring railway track gauge according to claim 2, wherein the actual point cloud computing step in the step 32 is as follows:
left and right rail inner side point set U of real side point cloud Q obtained through inner side face extraction ql ，U qr Plane fitting equations of S ql ，S qr For any point q and a certain plane S in the space, if the distance from the point q to the plane S is D (q, S), the track gauge D of the real point cloud is calculated according to the following formula q ：
Wherein n is l ，n r Respectively represent the collection U ql And U qr Scale of (2); the track gauge of the track is thus obtained.
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 CN111932676A (en) | 2020-11-13 |
| CN111932676B true 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) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN114332212B (en) * | 2022-03-11 | 2022-06-07 | 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 |
| CN115824070B (en) * | 2023-02-23 | 2023-05-30 | 湖南睿图智能科技有限公司 | Rail part size measurement method and device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9947128B2 (en) * | 2013-01-29 | 2018-04-17 | Andrew Robert Korb | Methods for improving accuracy, analyzing change detection, and performing data compression for 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 |
| CN108657222B (en) * | 2018-05-03 | 2019-06-07 | 西南交通大学 | Railroad track gauge and horizontal parameters measurement method based on vehicle-mounted Lidar 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 |
| CN110647798B (en) * | 2019-08-05 | 2023-01-03 | 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 |
| CN111429498B (en) * | 2020-03-26 | 2022-08-30 | 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
Also Published As
| Publication number | Publication date |
|---|---|
| CN111932676A (en) | 2020-11-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111932676B (en) | Railway track gauge rapid measurement device and method | |
| Liu et al. | A review of applications of visual inspection technology based on image processing in the railway industry | |
| Morgenthal et al. | Framework for automated UAS-based structural condition assessment of bridges | |
| Xie et al. | RRCNet: Rivet region classification network for rivet flush measurement based on 3-D point cloud | |
| CN111768417B (en) | Railway wagon overrun detection method based on monocular vision 3D reconstruction technology | |
| CN110866969A (en) | Engine blade reconstruction method based on neural network and point cloud registration | |
| CN112950532B (en) | Train pantograph state detection method | |
| CN103617328A (en) | Aircraft three-dimensional attitude calculation method | |
| US20220198695A1 (en) | Unmanned aerial vehicle platform based vision measurement method for static rigid object | |
| CN114037703B (en) | Subway valve state detection method based on two-dimensional positioning and three-dimensional attitude calculation | |
| Minghui et al. | Deep learning enabled localization for UAV autolanding | |
| CN108801136B (en) | Method for determining lateral displacement and attitude of vehicle model in vehicle stability experiment | |
| CN106886988A (en) | A kind of linear goal detection method and system based on unmanned aerial vehicle remote sensing | |
| CN110490342B (en) | Contact net static geometrical parameter detection method based on Faster R-CNN | |
| CN115857040A (en) | Dynamic visual detection device and method for foreign matters on locomotive roof | |
| CN113763423B (en) | Systematic target recognition and tracking method based on multi-mode data | |
| Wu et al. | Autonomous UAV landing system based on visual navigation | |
| Yang et al. | Discretization–filtering–reconstruction: railway detection in images for navigation of inspection UAV | |
| Oskouie et al. | A data quality-driven framework for asset condition assessment using LiDAR and image data | |
| CN112862862A (en) | Airplane autonomous oil receiving device based on artificial intelligence visual tracking and application method | |
| CN112560922A (en) | Vision-based foggy-day airplane autonomous landing method and system | |
| CN117152706A (en) | Aircraft runway accumulated water identification method, device and system | |
| CN117058366A (en) | Large aircraft large part point cloud semantic segmentation method based on pre-training large model | |
| 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 |
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 |