CN115824237B - Rail pavement recognition method and device - Google Patents
Rail pavement recognition method and device Download PDFInfo
- Publication number
- CN115824237B CN115824237B CN202211513288.3A CN202211513288A CN115824237B CN 115824237 B CN115824237 B CN 115824237B CN 202211513288 A CN202211513288 A CN 202211513288A CN 115824237 B CN115824237 B CN 115824237B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- track
- section
- road surface
- target
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The application relates to the technical field of image recognition, and provides a rail road surface recognition method and device. The method comprises the following steps: dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section; determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, so as to obtain the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector; and determining the track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine the track pavement according to the track pavement point clouds of the point cloud sections. The track pavement identification method provided by the embodiment of the application can improve the accuracy of track pavement identification.
Description
Technical Field
The application relates to the technical field of image recognition, in particular to a rail road surface recognition method and device.
Background
Currently, with the growth and flow of population, the track traffic task in urban traffic is more heavy, and the automatic control of the track vehicles, such as unmanned, has higher requirements. In realizing the automatic control of the railway vehicles, the autonomous identification of the road surface of the railway to identify the road surface and the road surface obstacle in the road surface of the railway is an important one.
In the related art, the road surface is generally identified by acquiring point cloud data of the road surface, and then adopting a similar plane grid method or a clustering-based ground segmentation method to the acquired point cloud data, wherein the road surface identification modes can obtain ideal ground identification effect under the condition of a ground platform and obvious obstacles. However, because the track pavement has a complex environment, such as a small hillside built on a gravel pile, the track pavement is positioned at a high point of the topography; a track crossing the road at the crossing is arranged at the periphery of the track; the method has various working conditions such as construction near the track, continuous pit on the road surface and the like, and under the conditions, the obtained point cloud data are uneven, large-scale scattered point cloud data exist, and the existing road surface identification mode is not suitable for denoising the large-scale scattered point cloud data, so that the road surface and road surface barriers in the road surface of the track cannot be well separated, and the identification accuracy of the road surface of the track is affected.
Disclosure of Invention
The present application is directed to solving at least one of the technical problems existing in the related art. Therefore, the application provides the track pavement identification method which can improve the accuracy of track pavement identification.
The application further provides a track pavement recognition device.
The application further provides electronic equipment.
The application also proposes a computer readable storage medium.
According to an embodiment of the first aspect of the application, a track pavement identification method comprises the following steps:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, so as to obtain the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
and determining the track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine the track pavement according to the track pavement point clouds of the point cloud sections.
According to the track pavement identification method provided by the embodiment of the application, the track pavement point cloud map of the track vehicle in the travelling direction is divided, after each point cloud section is obtained, the average pavement height and the target pavement normal vector of the point cloud section are determined according to the Z-axis coordinates of each target point in the point cloud section, after the target point cloud in the point cloud section is obtained according to the average pavement height and the target pavement normal vector, the track pavement point cloud of the point cloud section is determined according to the target point cloud, and then the track pavement is determined according to the track pavement point cloud of each point cloud section. The average road surface height and the target road surface normal vector obtained by the Z-axis coordinates of each target point in the point cloud section are utilized to independently calculate the track road surface point cloud of the point cloud section, and then the track road surface point cloud of each point cloud section is used to determine the final track road surface so as to reduce the influence of large-scale scattered point cloud on track road surface identification and further improve the accuracy of track road surface identification.
According to one embodiment of the present application, dividing a track point cloud map located in a traveling direction of a track vehicle to obtain each point cloud section includes:
and dividing the track point cloud map according to the preset length to obtain each point cloud section.
According to one embodiment of the present application, the track point cloud map is divided to obtain each point cloud section, including:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each initial section;
acquiring a track line section corresponding to the initial section from a track line map in the travelling direction of the railway vehicle according to the position information of the initial section;
and removing outliers which are not matched with the track line section from the initial section, and acquiring the point cloud section.
According to one embodiment of the present application, removing outliers from the initial section, which are not matched with the track line section, and obtaining the point cloud section includes:
converting the track line section into a vehicle coordinate system of the track vehicle to obtain a target line section;
and removing outliers which are not matched with the target line section from the initial section, and acquiring the point cloud section.
According to one embodiment of the present application, determining the average road surface height and the target road surface normal vector of the point cloud segment according to the Z-axis coordinates of each target point includes:
acquiring an average Z-axis coordinate according to the Z-axis coordinate of each target point, so as to determine the average road surface height according to the average Z-axis coordinate; the method comprises the steps of,
and carrying out PCA processing on the Z-axis coordinate of each target point to obtain the normal vector of the target road surface.
According to one embodiment of the present application, determining the track road surface point cloud of each of the point cloud sections according to the target point cloud of each of the point cloud sections includes:
marking the target point cloud of the point cloud section as the current target point cloud, carrying out pavement identification processing on the current target point cloud for a plurality of times, after determining the next target point cloud in the current pavement section according to the current pavement height and the current pavement normal vector of the current target point cloud, iterating the next target point cloud into the current target point cloud until the iteration times reach the preset times, and determining the current target point cloud obtained by the last iteration as the track pavement point cloud of the point cloud section;
and determining the current road surface height and the current road surface normal vector according to the Z-axis coordinate of each point data in the current target point cloud.
According to one embodiment of the present application, the target point is point data in which a coordinate value of a Z-axis coordinate in the point cloud section is smaller than a preset value.
According to a second aspect of the present application, a track pavement recognition device includes:
the point cloud map segmentation module is used for segmenting a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
the target point cloud acquisition module is used for determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section so as to acquire the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
the track pavement identification module is used for determining track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections so as to determine track pavement according to the track pavement point clouds of the point cloud sections.
An electronic device according to an embodiment of a third aspect of the present application includes a processor and a memory storing a computer program, the processor implementing the track road surface recognition method according to any of the above embodiments when executing the computer program.
A computer-readable storage medium according to an embodiment of a fourth aspect of the present application has stored thereon a computer program which, when executed by a processor, implements the track road surface recognition method according to any of the above-described embodiments.
A computer program product according to an embodiment of the fifth aspect of the present application comprises: the computer program when executed by a processor implements the rail road surface recognition method according to any one of the embodiments described above.
The above technical solutions in the embodiments of the present application have at least one of the following technical effects:
the method comprises the steps of dividing a track point cloud map in the travelling direction of a railway vehicle, obtaining each point cloud section, determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, obtaining the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector, determining the track road surface point cloud of the point cloud section according to the target point cloud, and determining the track road surface according to the track road surface point cloud of each point cloud section. The average road surface height and the target road surface normal vector obtained by the Z-axis coordinates of each target point in the point cloud section are utilized to independently calculate the track road surface point cloud of the point cloud section, and then the track road surface point cloud of each point cloud section is used to determine the final track road surface so as to reduce the influence of large-scale scattered point cloud on track road surface identification and further improve the accuracy of track road surface identification.
Drawings
In order to more clearly illustrate the application or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a track pavement recognition method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of further refining the acquisition of the point cloud segment in the track pavement identification method of FIG. 1 in an embodiment of the present application;
FIG. 3 is a schematic diagram of a relative rotation angle of a coordinate system according to an embodiment of the present application;
FIG. 4 is a schematic flow chart of further refining the determination of the track pavement in the track pavement identification method of FIG. 1 in an embodiment of the present application;
fig. 5 is a schematic structural diagram of a track pavement recognition device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The track pavement recognition method and device provided by the embodiment of the application are described and illustrated in detail below through several specific embodiments.
In one embodiment, a track pavement identification method is provided, and is applied to a server for identifying a track pavement. The server may be an independent server or a server cluster formed by a plurality of servers, and may also be a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, basic cloud computing services such as big data and artificial intelligent sampling point devices, and the like.
As shown in fig. 1, the track pavement identification method provided in this embodiment includes:
step 101, acquiring Z-axis coordinates of each target point from any point cloud section of a track point cloud map located in the travelling direction of a track vehicle;
102, determining an average road surface height and a target road surface normal vector of the point cloud section according to Z-axis coordinates of each target point, so as to obtain a target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
step 103, determining the track road surface point clouds of each point cloud section according to the target point clouds of each point cloud section.
The method comprises the steps of dividing a track point cloud map in the travelling direction of a railway vehicle, obtaining each point cloud section, determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, obtaining the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector, determining the track road surface point cloud of the point cloud section according to the target point cloud, and determining the track road surface according to the track road surface point cloud of each point cloud section. The average road surface height and the target road surface normal vector obtained by the Z-axis coordinates of each target point in the point cloud section are utilized to independently calculate the track road surface point cloud of the point cloud section, and then the track road surface point cloud of each point cloud section is used to determine the final track road surface so as to reduce the influence of large-scale scattered point cloud on track road surface identification and further improve the accuracy of track road surface identification.
In an embodiment, the track point cloud map in the traveling direction of the railway vehicle may be acquired by a laser radar mounted on the railway vehicle. Wherein the sensing part of the laser radar faces the travelling direction of the railway vehicle. The laser radar is a short term of laser detection and ranging system, and is a product of combining laser technology and radar technology. The laser radar adopts a laser as a radar of a radiation source, and generally consists of a transmitter, an antenna, a receiver, a tracking frame, information processing and the like. The transmitter is a laser of various forms; the antenna is an optical telescope; the receiver adopts various forms of optical point detectors; the laser radar adopts two working modes of pulse or continuous wave, and the detection method is divided into direct detection and heterodyne detection. The LIDAR system includes a single beam narrowband laser and a receiving system. The laser generates and emits a beam of light pulses that impinges on the object and reflects back to be received by the receiver. The receiver accurately measures the propagation time of the light pulse from the emission to the reflection back. Because the light pulse propagates at the speed of light, the receiver always receives the previous reflected pulse before the next pulse is sent out. In view of the fact that the speed of light is known, the travel time can be converted into a measure of distance. The coordinates X, Y and Z of each ground light spot can be accurately calculated by combining the height of the laser, the laser scanning angle, the position of the laser obtained from the GPS and the laser emitting direction obtained from the INS.
The frequency of the laser beam emission may range from a few pulses per second to tens of thousands of pulses per second. For example, a system with ten thousand pulses per second would have a receiver recording sixty-thousand points in one minute. In general, the ground spot spacing of LIDAR systems varies from 2-4 m. Lidar is a radar system operating in the infrared to ultraviolet spectrum and its principle and construction are very similar to that of a laser range finder. The laser radar has the function of accurately measuring the position (distance and angle), the motion state (speed, vibration and gesture) and the shape of a target, and detecting, identifying, distinguishing and tracking the target.
Since the rail vehicle must run on the rail, the laser radar carried on the rail vehicle can acquire the rail point cloud map including the point cloud of the rail.
After the track point cloud map is acquired, the track point cloud map can be segmented from far to near or from near to far in the travelling direction of the track vehicle to obtain a plurality of point cloud sections. For example, assuming that the traveling direction of the rail vehicle is the Y-axis direction, the rail point cloud map is cut multiple times in the Y-axis direction, thereby obtaining multiple point cloud sections with the same X-axis and Y-axis lengths.
When the track point cloud map is segmented, the track point cloud map can be segmented according to the preset length, and each point cloud section with the same length is obtained. Therefore, the quantity of scattered point clouds in each point cloud section is ensured to be approximately consistent as much as possible, and the defect that the consistency of the identification result of the point cloud section is insufficient due to the fact that the scattered point clouds of part of the point cloud sections are too many and the scattered point clouds of part of the point cloud sections are less due to the fact that the point cloud sections are not uniform is avoided.
In consideration of the acquired track point cloud map, the acquired point cloud section may include uneven ground and/or broken stone near the track, which results in that the acquired point cloud section includes not only the point cloud on the track line but also discrete points outside the track line, thereby affecting the accuracy of subsequent identification of the track pavement. Therefore, to avoid the influence of discrete points on the track pavement recognition, in an embodiment, as shown in fig. 2, the track point cloud map is divided to obtain each point cloud segment, which includes:
step 201, dividing a track point cloud map located in the travelling direction of a railway vehicle to obtain each initial section;
step 202, according to the position information of the initial section, acquiring a track line section corresponding to the initial section from a track line map in the travelling direction of the railway vehicle;
and 203, removing outliers which are not matched with the track line section from the initial section, and acquiring the point cloud section.
In an embodiment, after the track point cloud map is obtained, the track point cloud map is segmented according to a preset length to obtain a plurality of initial sections with consistent lengths. Then, from the vehicle positioning data of the rail vehicle, and the distance between each initial section and the rail vehicle, the position information of each initial section, such as the position information of 4 vertices of each initial section, can be determined. Meanwhile, a track route map in the traveling direction of the railway vehicle can be acquired from the high-precision track map according to the vehicle positioning data of the railway vehicle and the traveling direction of the railway vehicle. After the position information of the initial section is obtained, the track line section corresponding to the initial section can be obtained from the track line map according to the position information of the initial section, for example, a rectangular area surrounded by 4 vertexes is formed in the track line map according to the position information of the 4 vertexes of the initial section, and the track line in the rectangular area is the track line section corresponding to the initial section.
After the track line section is acquired, since the track line section is a two-dimensional image and the initial section is a three-dimensional image, the two-dimensional coordinates of each pixel point in the track line section can be matched with the three-dimensional coordinates of each point data in the initial section. If the abscissa and the ordinate of the three-dimensional coordinate of a certain point data in the initial section are not matched with the abscissa and the ordinate of the two-dimensional coordinate of each pixel point, the point is indicated to be an outlier, and the outlier is removed from the initial section. After all outliers are removed, a point cloud section can be obtained, so that when the track point cloud map is segmented, the interference of surrounding environments of the track vehicle is removed, and only the point cloud of the front track area of the track vehicle is reserved, so that the accuracy of the subsequent track pavement identification is improved.
To make outlier rejection more accurate to obtain a more accurate point cloud segment, in one embodiment, rejecting outliers from the initial segment that do not match the track line segment, the obtaining the point cloud segment includes:
converting the track line section into a vehicle coordinate system of the track vehicle to obtain a target line section;
and removing outliers which are not matched with the target line section from the point cloud data, and acquiring the point cloud section.
Since the track line map is usually generated based on a map coordinate system, that is, a geodetic coordinate system, and the track point cloud map is generated based on a vehicle coordinate system of the railway vehicle, in order to avoid inaccurate screening of outliers due to different coordinate systems, in one embodiment, after the track line section corresponding to the initial section is obtained from the track line map, coordinates of each pixel point in the track line section are converted into the vehicle coordinate system of the railway vehicle. The coordinate conversion of each pixel point may include:
coordinates of the front and rear wheels of the railway vehicle in the vehicle coordinate system OXY are acquired and are marked as (x 1, y 1), (x 2, y 2). And (3) acquiring coordinates of the front wheel and the rear wheel of the railway vehicle in an earth coordinate system, namely (X '1, Y' 1) and (X '2, Y' 2). The conversion relation of the two coordinate systems is obtained through the coordinates of the two points under different coordinate systems.
Specifically, the coordinates of the two coordinate systems, namely, the vehicle coordinate system OXY, the geodetic coordinate system O ' X ' Y ', and the geodetic coordinate system O ' X ' Y ' origin O ' under the vehicle coordinate system OXY are (X0, Y0), and the relative rotation angle of the two coordinate systems is θ as shown in fig. 3.
Two-point relationships are known: p1 (x 1, y 1) → (x '1, y' 1); p2 (X2, Y2) → (X '2, Y ' 2), the coordinates (X ', Y ') of any (X, Y) in O ' X ' Y ' are determined.
The conversion relation of the two coordinate systems can be obtained:
x′＝(x-x 0 )·cosθ+(y-y 0 )·sinθ
y′＝-(x-x 0 )·sinθ+(y-y 0 )·cosθ
after conversion, can be obtained:
x′＝x·cosθ+y·sinθ-x 0 ·cosθ-y 0 ·sinθ
y′＝-x·sinθ+y·cosθ+x 0 ·sinθ-y 0 ·cosθ
let a= -x 0 ·cosθ-y 0 ·sinθb＝x 0 ·sinθ-y 0 Cos θ available:
x′＝x·cosθ+y·sinθ+a
y′＝-x·sinθ+y·cosθ+b
substituting P1, P2 into the above formula, we can obtain:
x′＝x 1 ·cosθ+y 1 ·sinθ+a ①
y′＝-x 1 ·sinθ+y 1 ·cosθ+b ②
x′＝x 2 ·cosθ+y 2 ·sinθ+a ③
y′＝-x 2 ·sinθ+y 2 ·cosθ+b ④
the rotation angle θ is calculated, and the equation ((2) + (4))/(1) + (3)) can be obtained:
x′ 1 -x′ 2 ＝(x 1 -x 2 )·cosθ+(y 1 -y 2 )·sinθ ⑤
y′ 1 -y′ 2 ＝-(x 1 -x 2 )·sinθ+(y 1 -y 2 )·cosθ ⑥
and (6)/(5), can be obtained by:
according to the above formula:
((1) + (3))/2, ((2) + (4))/2 gives a, b:
b can be obtained:
at this time, the conversion relation of the two coordinate systems can be found, so that the track line section in the geodetic coordinate system can be converted into the vehicle coordinate system of the railway vehicle by the conversion relation to obtain the target line section.
After the target line section is obtained, the target line section and the initial section are in the same coordinate system, so that the target line section and the initial section can be matched at the moment, outliers which are not matched with the target line section can be removed from the initial section, and the point cloud section can be obtained.
After the point cloud segment is acquired, all point data can be extracted from the point cloud segment as a target point. In order to improve the accuracy of the subsequent track road identification, point data with the coordinate value of the Z-axis coordinate smaller than a preset value can be extracted from the point cloud section as target points, and each target point is a part of the point cloud with smaller height in the point cloud section, so that the probability of the point data of the track road is high.
After each target point of the point cloud section is extracted, the Z-axis coordinate of each target point is extracted, then the Z-axis coordinate of each target point is averaged, and after the average Z-axis coordinate of each target point is obtained, the average Z-axis coordinate is taken as the average road surface height. And PCA (Principal Components Analysis) processing is carried out on the Z-axis coordinates of each target point, so that the normal vector of the target road surface can be obtained.
Where PCA is the projection of three dimensions onto a surface for finding its dominant direction. The selection of the surface is based on selecting the surface which maximizes the distribution variance of the points as much as possible, and the larger the variance is, the larger the information amount thereof is. Since the large probability of each target point includes the point data of the track road surface, the PCA principal component analysis is performed on each target point, so that a vector perpendicular to the track ground, namely the normal vector of the target road surface, is obtained.
After the average road surface height and the target road surface normal vector are obtained, a Z-axis coordinate can be determined as the average road surface height, and a plane perpendicular to the target road surface normal vector is formed, and then all point data positioned in the plane in the point cloud section are determined as the target point cloud of the point cloud section. At this time, the point data except the point cloud in the point cloud section may be determined as the point data of the non-track road surface.
After the target point cloud of the point cloud section is obtained, the target point cloud of the point cloud section can be used as the track road surface point cloud of the point cloud section, and then the track road surface point cloud of the point cloud section is fitted to form the track road surface.
In order to make the obtained track road surface more accurate, in an embodiment, as shown in fig. 4, determining the track road surface point cloud of each point cloud section according to the target point cloud of each point cloud section includes:
step 301, marking a target point cloud of the point cloud section as a current target point cloud, performing pavement identification processing on the current target point cloud for a plurality of times, after determining a next target point cloud in the current pavement section according to a current pavement height and a current pavement normal vector of the current target point cloud, iterating the next target point cloud as the current target point cloud until the iteration number reaches a preset number, and determining the current target point cloud obtained in the last iteration as a track pavement point cloud of the point cloud section;
and determining the current road surface height and the current road surface normal vector according to the Z-axis coordinate of each point data in the current target point cloud.
In an embodiment, after a target point cloud of a point cloud section is obtained, the target point cloud is marked as a current target point cloud, then the Z-axis coordinates of each point data in the current target point cloud are averaged, the average Z-axis coordinates are obtained as a current road surface height, and meanwhile, the Z-axis coordinates of each point data in the current target point cloud are subjected to PCA processing to obtain a current road surface normal vector. After the current road surface height and the current road surface normal vector are obtained, a Z-axis coordinate can be determined as the current road surface height, and a plane perpendicular to the current road surface normal vector is formed, and then all point data positioned on the plane in the current target point cloud are combined into the next target point cloud. After the next target point cloud is obtained, iterating the next target point cloud into the current target point cloud, repeating the steps, iterating for N times, for example, after 5 times, determining the current target point cloud formed after the 5 th iteration as the track pavement point cloud of the point cloud section.
After the track pavement point clouds of the point cloud sections are determined in the mode, the track pavement point clouds of the point cloud sections can be fitted, and therefore the track pavement in the travelling direction of the track vehicle is obtained.
The obtained track road surface point clouds of the point cloud sections are more accurate in a multi-iteration mode, and the accuracy of the track road surface determined according to the track road surface point clouds is further improved.
The track road surface recognition device provided by the application is described below, and the track road surface recognition device described below and the track road surface recognition method described above can be referred to correspondingly.
In one embodiment, as shown in fig. 5, there is provided a track pavement recognition apparatus including:
the point cloud map segmentation module 210 is configured to segment a track point cloud map located in a traveling direction of the track vehicle, and obtain each point cloud segment;
a target point cloud obtaining module 220, configured to determine an average road surface height and a target road surface normal vector of the point cloud segment according to a Z-axis coordinate of each target point in the point cloud segment, so as to obtain a target point cloud in the point cloud segment according to the average road surface height and the target road surface normal vector;
the track road surface identification module 230 is configured to determine track road surface point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine a track road surface according to the track road surface point clouds of the point cloud sections.
The method comprises the steps of dividing a track point cloud map in the travelling direction of a railway vehicle, obtaining each point cloud section, determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, obtaining the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector, determining the track road surface point cloud of the point cloud section according to the target point cloud, and determining the track road surface according to the track road surface point cloud of each point cloud section. The average road surface height and the target road surface normal vector obtained by the Z-axis coordinates of each target point in the point cloud section are utilized to independently calculate the track road surface point cloud of the point cloud section, and then the track road surface point cloud of each point cloud section is used to determine the final track road surface so as to reduce the influence of large-scale scattered point cloud on track road surface identification and further improve the accuracy of track road surface identification.
In one embodiment, the point cloud map segmentation module 210 is specifically configured to:
and dividing the track point cloud map according to the preset length to obtain each point cloud section.
In one embodiment, the point cloud map segmentation module 210 is specifically configured to:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each initial section;
acquiring a track line section corresponding to the initial section from a track line map in the travelling direction of the railway vehicle according to the position information of the initial section;
and removing outliers which are not matched with the track line section from the initial section, and acquiring the point cloud section.
In one embodiment, the point cloud map segmentation module 210 is specifically configured to:
converting the track line section into a vehicle coordinate system of the track vehicle to obtain a target line section;
and removing outliers which are not matched with the target line section from the initial section, and acquiring the point cloud section.
In an embodiment, the target point cloud acquisition module 220 is specifically configured to:
acquiring an average Z-axis coordinate according to the Z-axis coordinate of each target point, so as to determine the average road surface height according to the average Z-axis coordinate; the method comprises the steps of,
and carrying out PCA processing on the Z-axis coordinate of each target point to obtain the normal vector of the target road surface.
In one embodiment, the track pavement identification module 230 is specifically configured to:
marking the target point cloud of the point cloud section as the current target point cloud, carrying out pavement identification processing on the current target point cloud for a plurality of times, after determining the next target point cloud in the current pavement section according to the current pavement height and the current pavement normal vector of the current target point cloud, iterating the next target point cloud into the current target point cloud until the iteration times reach the preset times, and determining the current target point cloud obtained by the last iteration as the track pavement point cloud of the point cloud section;
and determining the current road surface height and the current road surface normal vector according to the Z-axis coordinate of each point data in the current target point cloud.
In an embodiment, the target point is point data with a coordinate value of a Z-axis coordinate in the point cloud segment smaller than a preset value.
Fig. 6 illustrates a physical schematic diagram of an electronic device, as shown in fig. 6, which may include: processor 810, communication interface (Communication Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. The processor 810 may call a computer program in the memory 830 to perform a track road surface identification method, for example comprising:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, so as to obtain the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
and determining the track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine the track pavement according to the track pavement point clouds of the point cloud sections.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method of the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, an embodiment of the present application further provides a storage medium, where the storage medium includes a computer program, where the computer program may be stored on a non-transitory computer readable storage medium, and when the computer program is executed by a processor, the computer is capable of executing the track pavement recognition method provided in the foregoing embodiments, for example, including:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, so as to obtain the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
and determining the track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine the track pavement according to the track pavement point clouds of the point cloud sections.
In another aspect, an embodiment of the present application further provides a processor readable storage medium, where a computer program is stored, where the computer program is configured to cause a processor to perform a method provided in the foregoing embodiments, for example, including:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
determining the average road surface height and the target road surface normal vector of the point cloud section according to the Z-axis coordinates of each target point in the point cloud section, so as to obtain the target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
and determining the track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine the track pavement according to the track pavement point clouds of the point cloud sections.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (7)
1. A method of identifying a track pavement, comprising:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
acquiring an average Z-axis coordinate according to the Z-axis coordinate of each target point, determining an average road surface height according to the average Z-axis coordinate, performing PCA processing on the Z-axis coordinate of each target point, and acquiring a target road surface normal vector, so as to acquire a target point cloud in the point cloud section according to the average road surface height and the target road surface normal vector;
determining the track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections, so as to determine the track pavement according to the track pavement point clouds of the point cloud sections;
dividing the track point cloud map to obtain each point cloud section, including:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each initial section;
acquiring a track line section corresponding to the initial section from a track line map in the travelling direction of the railway vehicle according to the position information of the initial section;
removing outliers which are not matched with the track line section from the initial section, and obtaining the point cloud section;
determining the track road surface point cloud of each point cloud section according to the target point cloud of each point cloud section, including:
marking the target point cloud of the point cloud section as the current target point cloud, carrying out pavement identification processing on the current target point cloud for a plurality of times, after determining the next target point cloud in the current point cloud section according to the current pavement height and the current pavement normal vector of the current target point cloud, iterating the next target point cloud into the current target point cloud until the iteration times reach the preset times, and determining the current target point cloud obtained by the last iteration as the track pavement point cloud of the point cloud section;
and determining the current road surface height and the current road surface normal vector according to the Z-axis coordinate of each point data in the current target point cloud.
2. The track pavement recognition method according to claim 1, wherein dividing a track point cloud map located in a traveling direction of a track vehicle to obtain each point cloud section includes:
and dividing the track point cloud map according to the preset length to obtain each point cloud section.
3. The track pavement identification method of claim 1, wherein removing outliers from the initial segment that do not match the track line segment, obtaining the point cloud segment, comprises:
converting the track line section into a vehicle coordinate system of the track vehicle to obtain a target line section;
and removing outliers which are not matched with the target line section from the initial section, and acquiring the point cloud section.
4. The method according to claim 1, wherein the target point is point data in which a coordinate value of a Z-axis coordinate in the point cloud section is smaller than a preset value.
5. A track pavement identification device, comprising:
the point cloud map segmentation module is used for segmenting a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each point cloud section;
the target point cloud acquisition module is used for acquiring average Z-axis coordinates according to the Z-axis coordinates of each target point, determining average road surface height according to the average Z-axis coordinates, performing PCA (principal component analysis) processing on the Z-axis coordinates of each target point, and acquiring target road surface normal vectors, so as to acquire target point clouds in the point cloud section according to the average road surface height and the target road surface normal vectors;
the track pavement identification module is used for determining track pavement point clouds of the point cloud sections according to the target point clouds of the point cloud sections so as to determine track pavement according to the track pavement point clouds of the point cloud sections;
the point cloud map separation module is specifically configured to:
dividing a track point cloud map positioned in the travelling direction of the railway vehicle to obtain each initial section;
acquiring a track line section corresponding to the initial section from a track line map in the travelling direction of the railway vehicle according to the position information of the initial section;
removing outliers which are not matched with the track line section from the initial section, and obtaining the point cloud section;
the track pavement identification module is specifically used for:
marking the target point cloud of the point cloud section as the current target point cloud, carrying out pavement identification processing on the current target point cloud for a plurality of times, after determining the next target point cloud in the current point cloud section according to the current pavement height and the current pavement normal vector of the current target point cloud, iterating the next target point cloud into the current target point cloud until the iteration times reach the preset times, and determining the current target point cloud obtained by the last iteration as the track pavement point cloud of the point cloud section;
and determining the current road surface height and the current road surface normal vector according to the Z-axis coordinate of each point data in the current target point cloud.
6. An electronic device comprising a processor and a memory storing a computer program, characterized in that the processor implements the track road surface recognition method of any one of claims 1 to 4 when executing the computer program.
7. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the rail road surface identification method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211513288.3A CN115824237B (en) | 2022-11-29 | 2022-11-29 | Rail pavement recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211513288.3A CN115824237B (en) | 2022-11-29 | 2022-11-29 | Rail pavement recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115824237A CN115824237A (en) | 2023-03-21 |
| CN115824237B true CN115824237B (en) | 2023-09-26 |
Family
ID=85532754
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211513288.3A Active CN115824237B (en) | 2022-11-29 | 2022-11-29 | Rail pavement recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115824237B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845321A (en) * | 2015-12-03 | 2017-06-13 | 高德软件有限公司 | The treating method and apparatus of pavement markers information |
| WO2021134339A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳元戎启行科技有限公司 | Point cloud-based segmentation processing method and apparatus, computer device, and storage medium |
| CN113379923A (en) * | 2021-06-22 | 2021-09-10 | 北醒（北京）光子科技有限公司 | Track identification method, device, storage medium and equipment |
| CN113536883A (en) * | 2021-03-23 | 2021-10-22 | 长沙智能驾驶研究院有限公司 | Obstacle detection method, vehicle, device, and computer storage medium |
| CN113570665A (en) * | 2021-07-30 | 2021-10-29 | 苏州挚途科技有限公司 | Road edge extraction method and device and electronic equipment |
-
2022
- 2022-11-29 CN CN202211513288.3A patent/CN115824237B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845321A (en) * | 2015-12-03 | 2017-06-13 | 高德软件有限公司 | The treating method and apparatus of pavement markers information |
| WO2021134339A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳元戎启行科技有限公司 | Point cloud-based segmentation processing method and apparatus, computer device, and storage medium |
| CN113536883A (en) * | 2021-03-23 | 2021-10-22 | 长沙智能驾驶研究院有限公司 | Obstacle detection method, vehicle, device, and computer storage medium |
| CN113379923A (en) * | 2021-06-22 | 2021-09-10 | 北醒（北京）光子科技有限公司 | Track identification method, device, storage medium and equipment |
| CN113570665A (en) * | 2021-07-30 | 2021-10-29 | 苏州挚途科技有限公司 | Road edge extraction method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115824237A (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109791052B (en) | Method and system for classifying data points of point cloud by using digital map | |
| CN109521756B (en) | Obstacle motion information generation method and apparatus for unmanned vehicle | |
| CN107850449B (en) | Method and system for generating and using positioning reference data | |
| Hata et al. | Feature detection for vehicle localization in urban environments using a multilayer LIDAR | |
| Chen et al. | Next generation map making: Geo-referenced ground-level LIDAR point clouds for automatic retro-reflective road feature extraction | |
| CN110794406B (en) | Multi-source sensor data fusion system and method | |
| CN105158762A (en) | Identifying and tracking convective weather cells | |
| Gikas et al. | A novel geodetic engineering method for accurate and automated road/railway centerline geometry extraction based on the bearing diagram and fractal behavior | |
| CN112781599B (en) | Method for determining the position of a vehicle | |
| CN112673280A (en) | Road detection method for a motor vehicle equipped with a LIDAR sensor | |
| CN113835102B (en) | Lane line generation method and device | |
| CN112379393B (en) | Train collision early warning method and device | |
| CN111999744A (en) | Unmanned aerial vehicle multi-azimuth detection and multi-angle intelligent obstacle avoidance method | |
| CN113376643A (en) | Distance detection method and device and electronic equipment | |
| CN115824237B (en) | Rail pavement recognition method and device | |
| WO2020135325A1 (en) | Mobile device positioning method, device and system, and mobile device | |
| CN113895482B (en) | Train speed measuring method and device based on trackside equipment | |
| Hofmann et al. | Quality assessment of automatically generated feature maps for future driver assistance systems | |
| CN114119465A (en) | Point cloud data processing method and device | |
| Sun et al. | Mathematical Method for Lidar-based Obstacle Detection of Intelligent Vehicle | |
| Huang et al. | Ground filtering algorithm for mobile LIDAR using order and neighborhood point information | |
| Huang et al. | An efficient multi-threshold selection method for lane detection based on lidar | |
| Farrell et al. | Precision mapping of the california connected vehicle testbed corridor | |
| Bai | Accurate obstacle prediction method in unmanned vehicle driving. | |
| JP7308772B2 (en) | DATA PROCESSING DEVICE, DATA PROCESSING METHOD AND DATA PROCESSING PROGRAM |
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 |