CN114187537B - Railway track and midline extraction method based on aerial remote sensing image - Google Patents
Railway track and midline extraction method based on aerial remote sensing image Download PDFInfo
- Publication number
- CN114187537B CN114187537B CN202111550857.7A CN202111550857A CN114187537B CN 114187537 B CN114187537 B CN 114187537B CN 202111550857 A CN202111550857 A CN 202111550857A CN 114187537 B CN114187537 B CN 114187537B
- Authority
- CN
- China
- Prior art keywords
- railway
- image
- line
- track
- aviation
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000005070 sampling Methods 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 238000013461 design Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 abstract description 5
- 238000005259 measurement Methods 0.000 abstract description 5
- 238000010276 construction Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000013102 re-test Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a railway track and midline extraction method based on aerial remote sensing images. The prior art is mainly completed by a manual total station and a laser scanner, the manual measurement has inefficiency, and the laser scanner has high cost and occupies railway operation. According to the method, image control point distribution is designed according to railway trend, aviation coverage images are obtained, and space three calculation is carried out on the aviation coverage images; extracting line segment information of each aviation coverage image by using a line segment extraction algorithm; drawing the same name trajectories from the image pairs, and recovering the space position information of the real space lines; and finally, track center line extraction and railway line information database construction are completed. The invention adopts a non-contact type space position sampling method, does not need to use a rail car for carrying out, does not influence normal railway operation activities in the data acquisition process, and has the characteristics of high efficiency, low cost and high operation safety.
Description
Technical Field
The invention belongs to the technical field of railway surveying, and particularly relates to a railway track and midline extraction method based on aerial remote sensing images.
Background
The existing track line reconstruction, retesting and checking technology is a conforming technology for railway design and construction, is a reconstruction and retesting technology for existing line design line data, can be even used as a foundation stability detection technology, and plays an important role in railway construction and operation management. The traditional method adopts an artificial line measurement or semi-automatic scanning mode, and adopts equipment such as a total station, a laser scanner and the like, however, the method using the total station can only measure line parameters, recover design geometric nodes and curvature, the sampling is insufficient, the space form of the existing railway cannot be truly expressed, and the linear fitness is relatively low; in order to solve the problem caused by insufficient sampling, the follow-up method adopts a laser scanner mode, and the spatial position accuracy of 2cm can be realized through the control of a POS system and a target along the line, but the method not only needs expensive laser scanning equipment and requires high-accuracy measurement efficiency of over millions of times per second, but also occupies normal railway running time and space, and influences economic activities such as normal railway operation.
The invention of patent CN110728689A discloses an existing railway central line extraction method based on laser point cloud data, and a binarization result is utilized to calculate a railway central line, but the method needs high-quality point cloud, relies on the effects of point cloud filtering and point cloud reflection intensity, and has serious prior knowledge constraint; the patent CN105844995A proposes a railway line operation maintenance measurement method of the vehicle-mounted LiDAR, which relies on the effect of classifying the rail surface data, and only needs to process components with complex fork structures separately to extract and form a line element information table, so as to complete a railway plane-longitudinal diagram.
Disclosure of Invention
In order to make up for the defects of the prior art, the invention provides a railway track and midline extraction method based on aerial remote sensing images, which solves the problems that the existing railway track and midline extraction technology cannot be compatible with high efficiency and economy, high precision and reliability and low cost.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
A railway track and midline extraction method based on aerial remote sensing images comprises the following steps:
step one: designing image control points according to the trend of the railway, wherein the image control points are distributed in pairs on two sides of the railway line;
Step two: acquiring an aviation coverage image of the image control point, wherein the overlapping degree of the aviation coverage image is more than or equal to 3 degrees;
step three: 2 routes with baseline lengths are externally expanded outside the range of the railway line area;
step four: performing space three-dimensional calculation on the aviation coverage images to obtain three-dimensional attitude information of each aviation coverage image in an object space and a camera distortion parameter model coefficient;
step five: extracting line segment information of each aviation coverage image by using a line segment extraction algorithm, and taking the line segment information as a prediction candidate value for predicting and extracting a post-acceleration semi-automatic trajectory;
Step six: obtaining an image pair with a spatial relationship, drawing a homonymy trajectory from the image pair in a semi-automatic tracking way, and then recovering the spatial position information of a real line in space in a least square homonymy trajectory fitting way;
Step seven: extracting a track center line;
step eight: and constructing a railway line information database by combining design standard parameters.
Specifically, the track centerline extraction includes the steps of:
step one: acquiring left and right orbit reference lines of the trajectories according to the positions of the same-name trajectories and the spatial position information of the spatial real lines, which are obtained in the step six;
step two: according to the left and right track reference lines, calculating a sampling line segment of a track center line according to a principle of normal vector intersection;
Step three: and directly fitting the parameters of the track out according to the sampling line segments to realize the reconstruction of the track surface information.
The invention has the beneficial effects that:
1) The invention adopts a non-contact type space position sampling method, does not need to use a rail car for carrying out, does not influence normal railway operation activities in the data acquisition process, and has important significance in retesting the existing railway lines;
2) The invention takes the photogrammetry technology as the core, the space measurement precision is consistent with the aviation mapping precision, the line mapping precision of 1:500 is satisfied, and the precision can be improved to millimeter level by improving the resolution;
3) The invention has the characteristics of high efficiency, low cost and high operation safety, does not need complex parameter setting and subsequent post-treatment, has controllable precision, and is beneficial to reducing capital investment in railway operation and maintenance;
4) The method does not need to set the threshold value by personnel with priori knowledge, classifies and filters redundant ground point information, does not need to input a large amount of labor cost, and can reconstruct and acquire the space position information of the railway track and the original parameters of the design.
Drawings
FIG. 1 is a flow chart of the operation of the present invention;
FIG. 2 is a diagram of the layout of control points along a railway;
FIG. 3 is a schematic diagram of aerial homonymy extraction;
Fig. 4 is a diagram of a railroad track isonym trace.
Detailed Description
The present invention will be described in detail with reference to the following embodiments.
The invention utilizes the current mature unmanned aerial vehicle aerial photography remote sensing mode with high cost performance and reliable precision to efficiently realize a non-contact line reconstruction method, which not only can check the line construction precision, but also can be used as a reliable existing railway central line reconstruction means. As shown in fig. 1, the present invention specifically includes the following steps:
Step one: designing image control point distribution according to railway trend, wherein the image control point distribution density, the resolution of images and parameters of cameras are required to be set according to project achievement precision requirements; as shown in fig. 2, the image control points are arranged in pairs at two sides of the railway line;
Step two: acquiring an aviation coverage image of an image control point, wherein the overlapping degree of the aviation coverage image is more than or equal to 3 degrees;
Step three: in order to meet the stability of the space three calculation, the airlines with 2 baseline lengths are required to be laid out outside the range of the railway line area, and the stability of the long strip area is required to be increased by the design thought of constructing the airlines;
Step four: performing air-to-air three-dimensional calculation on the aviation coverage image, and calculating three-dimensional attitude information of each image in an object space and a camera distortion parameter model coefficient according to the acquired aviation image, POS information and image control point information;
step five: extracting line segment information of each aviation coverage image by using a line segment extraction algorithm, and taking the line segment information as a prediction candidate value for predicting and extracting a post-acceleration semi-automatic trajectory;
Step six: obtaining an image pair with a spatial relationship, semi-automatically tracking and drawing a homonymy trajectory from the image pair, and then restoring the spatial position information of a real line in space in a least square homonymy trajectory fitting mode;
1) Since the same-name trajectories have epipolar constraint relationships in different images, the same-name trajectories must have epipolar constraint relationships in different images, as shown in fig. 3, the relationship is a constraint relationship of a basic matrix F in computer vision, and the specific presentation mode is shown in formula (1), wherein x1 and x2 are same-name points in left and right images, such as the same-name points in the trajectories, and the basic matrix F can be obtained through an inner parameter K and an outer parameter m= [ R T ] of left and right stereoscopic images, as shown in formula (2).
2) Besides the space geometric constraint existing in the stereopair, the track line in a single image also presents the characteristics of extremely strong smoothness and vividness, so that the image gray tracking track algorithm is another one-dimensional constraint of the track;
3) Combining 1) and 2), as shown in fig. 4, semi-automatic updating along the track line is realized by utilizing the constraint of the one-dimensional gray track and the space basic matrix;
4) Because of the regular nature of the orbit itself, the spatial constraint in 1) can be regarded as a line segment constraint, and the epipolar equation of the left image in the right image is Then selecting the edge line segment of the nearest neighborhood in the right image;
5) Space trajectory fitting, namely, drawing a semi-automatic trajectory through the steps 1) to 4), extracting pixel line segments from each image, wherein the pixel line segments are only required to be subjected to space multi-surface space intersection fitting in a segmented mode to obtain an optimal fitting line, and further, the flatness and smoothness constraint of the whole line segments are required to be considered, and a first derivative smoothness constraint is required to be added in a fitting function, and a specific fitting energy function is shown as a formula 2, wherein C i is a space surface corresponding to a line segment with the same name in an image i, a first term formula in a formula (3) is a fitting intersection formula, and a second term formula is used for ensuring that the fitted space curve has smooth characteristics;
Step seven: track centerline extraction
1) Acquiring left and right orbit reference lines of the trajectories according to the positions of the same-name trajectories and the spatial position information of the spatial real lines, which are obtained in the step six;
2) According to the left and right track reference lines, searching for a normal line which is perpendicular to the left and right aisle plane reference lines at the same time by a principle of normal vector intersection, and acquiring a midpoint of a left and right intersection point, namely a sampling point of a railway center line;
3) And directly fitting the parameters of the track out according to the sampling line segments to realize the reconstruction of the track surface information.
The obtained sampling points are all the supersampling expressions of the real rail surface, and a large amount of redundant information exists, so that the redundant information cannot be directly stored in a corresponding design database, and therefore the sampling data are required to be combined in the axial direction; the basis for eliminating redundant data is whether adjacent sampling points and lines are coaxial or not;
Step eight: and recording track parameters by combining design standard parameters such as track gauge, track height, waist thickness and the like in the track, and constructing a railway line information database.
Examples of applications are as follows:
If a certain section of operation railway in a hilly area is about 10km in length, runs in the north-south direction, has bridge roadbed and common roadbed sections along the way, directly utilizes the existing CPI control network as a known control network through collecting and early-stage preboring, then establishes regional image control points, wherein the absolute precision of a plane is better than 2cm, the absolute precision of a vertical plane is better than 3cm, the image control points are arranged in pairs along two sides of a railway line, and a pair of image control points are arranged in average 8 baseline lengths, and the pair is about 200 meters. The quality of the aviation image data is required to be clear, the color is saturated, and the ground resolution is set to be better than 1cm. The real attitude data of the high-precision image space can be obtained through space three-dimensional calculation, the error in the absolute plane position of the rail surface can be less than or equal to 2cm, the error in the absolute elevation position can be less than or equal to 3cm, and the measurement precision of millimeter level can be achieved if the ground resolution and the control point precision are improved. And comparing the resolving precision with the existing railway data, wherein the precision meets the retest requirement.
In the description of the present invention, unless explicitly stated and limited otherwise, the terms "disposed," "mounted," "connected," and "secured" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two elements or the interaction relationship of the two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
The content of the invention is not limited to the examples listed, and any equivalent transformation to the technical solution of the invention that a person skilled in the art can take on by reading the description of the invention is covered by the claims of the invention.
Claims (1)
1. The method for extracting the railway track and the central line based on the aerial remote sensing image is characterized by comprising the following steps of: the method comprises the following steps:
step one: designing image control points according to the trend of the railway, wherein the image control points are distributed in pairs on two sides of the railway line;
Step two: acquiring an aviation coverage image of the image control point, wherein the overlapping degree of the aviation coverage image is more than or equal to 3 degrees;
step three: 2 routes with baseline lengths are externally expanded outside the range of the railway line area;
step four: performing space three-dimensional calculation on the aviation coverage images to obtain three-dimensional attitude information of each aviation coverage image in an object space and a camera distortion parameter model coefficient;
step five: extracting line segment information of each aviation coverage image by using a line segment extraction algorithm, and taking the line segment information as a prediction candidate value for predicting and extracting a post-acceleration semi-automatic trajectory;
Step six: obtaining an image pair with a spatial relationship, drawing a homonymy trajectory from the image pair in a semi-automatic tracking way, and then recovering the spatial position information of a real line in space in a least square homonymy trajectory fitting way;
Step seven: extracting a track center line;
step eight: combining design standard parameters to construct a railway line information database;
The track centerline extraction comprises the following steps:
step one: acquiring left and right orbit reference lines of the trajectories according to the positions of the same-name trajectories and the spatial position information of the spatial real lines, which are obtained in the step six;
step two: according to the left and right track reference lines, calculating a sampling line segment of a track center line according to a principle of normal vector intersection;
Step three: and directly fitting the parameters of the track out according to the sampling line segments to realize the reconstruction of the track surface information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111550857.7A CN114187537B (en) | 2021-12-17 | 2021-12-17 | Railway track and midline extraction method based on aerial remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111550857.7A CN114187537B (en) | 2021-12-17 | 2021-12-17 | Railway track and midline extraction method based on aerial remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114187537A CN114187537A (en) | 2022-03-15 |
| CN114187537B true CN114187537B (en) | 2024-04-30 |
Family
ID=80544348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111550857.7A Active CN114187537B (en) | 2021-12-17 | 2021-12-17 | Railway track and midline extraction method based on aerial remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114187537B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100653627B1 (en) * | 2006-01-27 | 2006-12-05 | (자)한진개발공사 | Method for three-dimensional determining of basic design road route using three-dimensional geological spatial information treatment and aerial photograph |
| CN105844995A (en) * | 2016-05-20 | 2016-08-10 | 中铁第勘察设计院集团有限公司 | Railway line operation and maintenance measuring method based on vehicle-mounted LiDAR technology |
| CN106087621A (en) * | 2016-05-31 | 2016-11-09 | 中铁第四勘察设计院集团有限公司 | A kind of Existing Railway Line repetition measurement method based on mobile lidar technology |
| CN108763575A (en) * | 2018-06-06 | 2018-11-06 | 湖南省第测绘院 | Photo control point automatically selecting method based on photo control point database |
| RU2726256C1 (en) * | 2020-03-01 | 2020-07-10 | Дмитрий Александрович Рощин | Method of constructing three-dimensional model of terrain along railway track bed |
| CN113155098A (en) * | 2021-04-20 | 2021-07-23 | 中国铁路设计集团有限公司 | Existing railway track line high-precision three-dimensional reconstruction method based on unmanned aerial vehicle multi-view images |
-
2021
- 2021-12-17 CN CN202111550857.7A patent/CN114187537B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100653627B1 (en) * | 2006-01-27 | 2006-12-05 | (자)한진개발공사 | Method for three-dimensional determining of basic design road route using three-dimensional geological spatial information treatment and aerial photograph |
| CN105844995A (en) * | 2016-05-20 | 2016-08-10 | 中铁第勘察设计院集团有限公司 | Railway line operation and maintenance measuring method based on vehicle-mounted LiDAR technology |
| CN106087621A (en) * | 2016-05-31 | 2016-11-09 | 中铁第四勘察设计院集团有限公司 | A kind of Existing Railway Line repetition measurement method based on mobile lidar technology |
| CN108763575A (en) * | 2018-06-06 | 2018-11-06 | 湖南省第测绘院 | Photo control point automatically selecting method based on photo control point database |
| RU2726256C1 (en) * | 2020-03-01 | 2020-07-10 | Дмитрий Александрович Рощин | Method of constructing three-dimensional model of terrain along railway track bed |
| CN113155098A (en) * | 2021-04-20 | 2021-07-23 | 中国铁路设计集团有限公司 | Existing railway track line high-precision three-dimensional reconstruction method based on unmanned aerial vehicle multi-view images |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114187537A (en) | 2022-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110647798B (en) | Automatic track center line detection method based on vehicle-mounted mobile laser point cloud | |
| WO2023019709A1 (en) | Automatic detection method of conductor height and pull-out value of overhead line system based on vehicle-mounted mobile laser point cloud | |
| CN111429498B (en) | Railway business line three-dimensional center line manufacturing method based on point cloud and image fusion technology | |
| CN111045017B (en) | Method for constructing transformer substation map of inspection robot by fusing laser and vision | |
| Wang et al. | A point cloud-based robust road curb detection and tracking method | |
| CN112414309B (en) | High-speed rail contact line height-guiding and pull-out value inspection method based on airborne laser radar | |
| Papasaika et al. | Fusion of digital elevation models using sparse representations | |
| CN113125444A (en) | Railway bridge disease monitoring method based on unmanned aerial vehicle carrying three-dimensional scanner | |
| WO2020083103A1 (en) | Vehicle positioning method based on deep neural network image recognition | |
| CN109100719B (en) | Terrain map joint mapping method based on satellite-borne SAR (synthetic aperture radar) image and optical image | |
| CN105956542B (en) | High-resolution remote sensing image road extraction method based on statistical matching of structural wire harnesses | |
| CN111768417A (en) | Railway wagon overrun detection method based on monocular vision 3D reconstruction technology | |
| CN109544607A (en) | A kind of cloud data registration method based on road mark line | |
| CN103854290A (en) | Extended target tracking method based on combination of skeleton characteristic points and distribution field descriptors | |
| CN110926417B (en) | Vehicle-mounted railway tunnel detection system based on machine vision | |
| CN114299160A (en) | Dynamic detection method for pantograph running state based on fusion of vision and laser radar | |
| CN110490342B (en) | Contact net static geometrical parameter detection method based on Faster R-CNN | |
| CN104729529A (en) | Method and system for judging errors of topographic map surveying system | |
| Neubert et al. | Extraction of railroad objects from very high resolution helicopter-borne LiDAR and ortho-image data | |
| CN114897777A (en) | Full-automatic extraction method of laser point cloud of overhead line system supporting facility considering spatial relationship | |
| CN114187537B (en) | Railway track and midline extraction method based on aerial remote sensing image | |
| Kremer et al. | The RailMapper–A dedicated mobile LiDAR mapping system for railway networks | |
| CN107024175A (en) | Motorbus vehicle body critical size detecting system solution based on multi-vision visual | |
| CN111473774A (en) | Unmanned aerial vehicle single-flight-zone oblique photography power line obstacle inspection method | |
| CN106952249A (en) | Insulator chain axis detection method based on Cross ration invariability |
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 |