CN114494301A - Railway scene point cloud segmentation method based on airborne radar point cloud - Google Patents
Railway scene point cloud segmentation method based on airborne radar point cloud Download PDFInfo
- Publication number
- CN114494301A CN114494301A CN202210133610.3A CN202210133610A CN114494301A CN 114494301 A CN114494301 A CN 114494301A CN 202210133610 A CN202210133610 A CN 202210133610A CN 114494301 A CN114494301 A CN 114494301A
- Authority
- CN
- China
- Prior art keywords
- point
- point cloud
- initial seed
- specific sequence
- points
- 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.)
- Pending
Links
- 230000011218 segmentation Effects 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000001914 filtration Methods 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- 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
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20156—Automatic seed setting
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention relates to a point cloud segmentation method for a railway scene based on airborne radar point cloud, which comprises the following steps: s1, determining an initial seed point, and marking the initial seed point into a specific sequence; s2, searching a near-adjacent point, and adding the near-adjacent point into a current area if an included angle between the near-adjacent point and a normal of the seed point is smaller than a threshold value; s3, calculating a curvature value of a neighboring point, and if the curvature value is smaller than a threshold value, adding the curvature value into the specific sequence; s4, deleting the corresponding initial seed point; s5, repeating the steps S1-S4 until the initial seed point in the specific sequence is empty; s6, carrying out iterative processing on the specific sequence according to the steps S2-S5 until all points are marked into different areas; and S7, obtaining the target object. Each single-target object can be effectively segmented, different parameter ranges are provided under different environments, and selection under the condition of processing different railway scenes is facilitated.
Description
Technical Field
The invention relates to the technical field of railway detection, in particular to a railway scene point cloud segmentation method based on airborne radar point cloud.
Background
The railway construction in China is developed rapidly, the workload of daily management, maintenance and extension of the railway is increased greatly, and a large amount of capital and manpower is needed every year. At present, the survey of railways in China mainly adopts a contact type traditional measuring means. The traditional measuring technology mainly utilizes measuring instruments such as a level gauge, a total station, a gauging rule and a steel ruler, and combines special rail inspection vehicles and other tools to carry out, but the traditional measuring means is not only large in workload, low in efficiency, single in measurement result information and low in automation degree, but also has certain personal safety problem in the operation process of workers, and the traditional measuring means is difficult to adapt to the survey requirement of the current social development on the railway.
The three-dimensional laser scanning technology is a brand-new high-precision and high-efficiency data acquisition method and has the advantages of non-contact property, authenticity, rapidness, information comprehensiveness and the like. The three-dimensional laser scanning technology is introduced into the field of railway survey, comprehensive and abundant basic data can be provided for the railway survey through the acquisition of continuous point cloud data, the problem that the space state change condition of railway equipment along the line cannot be accurately and comprehensively reflected due to the fact that the data acquired by the traditional measuring means are discrete is solved, and support can be provided for maintenance, expansion, information management and the like of the railway. The airborne radar can rapidly acquire high-precision three-dimensional earth surface point cloud data of 0.1-1 m in a large area and derive digital products such as DSM (digital surface model) and DTM (digital time difference meter) and the like, the airborne radar can expand the acquisition range of the point cloud data and accelerate the acquisition rate, and meanwhile, massive point cloud data also provide requirements for the intelligent processing technology of the railway point cloud after the railway point cloud data are rapidly acquired. In the railway field, the existing mature and complete method for automatically extracting the characteristic information of points, lines, surfaces and the like in the railway point cloud data is relatively few, and still mainly depends on manual participation. Therefore, the research on how to apply the airborne radar technology to the field of railway survey and automatically extracting characteristic information such as railway line parameters from the point cloud data has important practical significance on state detection and daily maintenance of the line.
The development of the existing radar point cloud segmentation technology has also been greatly developed. And if the proximity relation and the r radius point density are introduced during extraction, the shelter seed points along the railway are well extracted, and then the segmentation is carried out by a region growing algorithm based on principal component analysis. The method can better segment and extract buildings, but has poor segmentation effect on vegetation with high density. The original three-dimensional discrete points need to be segmented, then segmented, and non-building areas are eliminated by setting an area threshold, and the method has a good segmentation effect on the premise of correct segmentation. The prior art often incorrectly segments a part of vegetation points into a point cloud set of a building in segmentation, and may cause some small objects to be incorrectly removed if an area threshold is not selected correctly. In summary, at the present stage, there is no point cloud segmentation method of the laser radar technology that can minimize the influence of incomplete segmentation.
However, the above prior art solutions have the following disadvantages: 1. the existing railway detection using the unmanned aerial vehicle is mainly based on the fact that the unmanned aerial vehicle acquires images, so that great accidental errors exist, and the requirements of real analysis of large data cannot be met by processing the increasingly complex railway line environment with two-dimensional images. Although the two-dimensional image can be used for identification, the problems that the size of the target cannot be judged, the length of a monitoring line is limited, the target is shielded and the like occur. 2. Airborne LiDAR point cloud data processing mainly relies on professional processing software, but if the railway topography that needs to measure is complicated, if only according to the automatic segmentation of parameter setting on a large scale, follow-up manual assistance cuts apart work load too big, and work efficiency is too low. 3. In the prior art, wrong segmentation exists, and in some railway scenes, small objects can be removed and objects which are too dense can not be effectively segmented due to the fact that the objects are similar to tall trees and too many interferents such as buildings. How to overcome the defects of the prior art scheme, a process of a segmentation method of railway point cloud acquired by an airborne radar is determined, and a method most suitable for railway point cloud segmentation is selected according to the characteristics of railway scene point cloud in a plurality of point cloud segmentation methods. The method overcomes the defect that the existing point cloud segmentation problem mainly depends on manual processing and has long processing time for the characteristic of large quantity of railway targets, realizes an automatic segmentation method for segmenting the railway scene point cloud target, and becomes a problem to be solved urgently in the technical field.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a railway scene point cloud segmentation method based on airborne radar point cloud, which specifically adopts the following technical scheme:
a point cloud segmentation method for a railway scene based on airborne radar point cloud comprises the following steps:
s1, determining an initial seed point, and marking the initial seed point into a specific sequence;
s2, searching a near-adjacent point, and adding the near-adjacent point into a current area if an included angle between the near-adjacent point and a normal of the seed point is smaller than a threshold value;
s3, calculating a curvature value of a neighboring point, and if the curvature value is smaller than a threshold value, adding the curvature value into the specific sequence;
s4, deleting the corresponding initial seed point;
s5, repeating the steps S1-S4 until the initial seed point in the specific sequence is empty;
s6, carrying out iterative processing on the specific sequence according to the steps S2-S5 until all points are marked into different areas;
and S7, obtaining the target object.
Further, the step S1 specifically includes: and finding the point with the minimum curvature value from the flattest area as the initial seed point, wherein the specific sequence is a sequence formed by the initial seed points.
Further, the step S2 specifically includes: searching m neighbor points which are closest to the Euclidean distance of one initial seed point, calculating the normal direction of each neighbor point and the normal direction of the corresponding initial seed point, and adding the neighbor point to the current area if the included angle between the normal of the neighbor point and the normal of the initial seed point is smaller than a normal included angle threshold value.
Further, the step S3 specifically includes: and calculating curvature values of the m adjacent points of the initial seed point, if the curvature values of the m adjacent points are smaller than a curvature threshold value, considering that the m adjacent points belong to the same region and combining the m adjacent points to obtain a combined seed point, and adding the combined seed point into the specific sequence.
Further, the step S6 specifically includes: and performing iterative processing on the combined seed points in the specific sequence according to the steps S2-S5 until all the points are marked into different areas.
Further, before the step S1, the method further includes: and acquiring point cloud data, and performing secondary filtering by adopting direct filtering and statistical outlier filtering to obtain complete point cloud data.
Further, the acquiring point cloud data specifically includes: using an unmanned aerial vehicle to carry out on-site line patrol flight, carrying out laser radar-carrying flight with the line overlapping degree not less than two thirds according to a preset route, and simultaneously acquiring LIDAR data and image data by using the laser radar; and transmitting the data to a base station, and performing combined solution of GNSS and IMU by the base station to obtain point cloud data.
Further, the threshold value of the normal included angle is 1-10 degrees.
Further, the curvature threshold value is 0.10-0.20.
Further, the normal included angle threshold is 5 °, and the curvature threshold is 0.12.
The technical scheme of the invention obtains the following beneficial effects: the defects that the traditional manual point cloud segmentation is complex in processing and needs a large amount of time and cost are overcome, the parameter range is provided for railway point cloud segmentation, and the test progress is accelerated. By a point cloud segmentation method similar to region growing, each single-target object can be segmented effectively, different parameter ranges are provided under different environments, and selection under the condition of processing different railway scenes is facilitated.
Drawings
FIG. 1 is an overall flow chart of the method of the present invention.
FIG. 2 is a flow chart of point cloud data sorting according to the present invention.
FIG. 3 is a complete point cloud data diagram of the present invention.
FIG. 4 is a point cloud segmentation result diagram according to the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby. It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The invention provides a railway scene point cloud segmentation method based on airborne laser radar data.
The specific embodiment of the invention takes the division of a tunnel portal in a railway scene as an example. The specific embodiment of the invention is suitable for the railway scene point cloud segmentation method after airborne radar point cloud data acquisition. A method for gathering pixel points based on the similarity of pixels in the same object region, starting from an initial region, incorporates adjacent or other regions of the same nature into the current region to gradually grow the region until there are no mergeable points or other small regions. The point cloud segmentation method is characterized in that the directions of normal vectors of the surfaces of objects are estimated and counted, whether the surfaces belong to the same object or not is judged according to the directions of the normal vectors, and finally segmentation of the three-dimensional point cloud is completed. The method can effectively avoid the problems of excessive segmentation and insufficient segmentation of the three-dimensional point cloud.
Referring to fig. 1, the specific algorithm flow of the embodiment of the present invention is as follows:
and searching a point with the minimum curvature value from the flattest area as an initial seed point, marking the area where the initial seed point is located, and adding a specific sequence. Each seed point forms an area after the addition of the adjacent points, and all the seed points are put into a specific sequence, so that the specific sequence is the set of each segmentation part in the point cloud cluster. The point where the curvature value is the smallest is located in the flattest area, and growing from the flattest area may reduce the total number of areas.
Searching a set of m neighbor points with the Euclidean distance to the seed point being the closest, calculating the normal direction of each neighbor point and the normal direction of the seed point, and adding the neighbor point to the current area if the included angle between the normal of the neighbor point and the normal of the seed point is less than a threshold value.
The specific sequence is initially composed of initial seed points. Calculating curvature values of m adjacent points of the initial seed point, if the curvature values of the m adjacent points are smaller than a threshold value, regarding the m adjacent points as belonging to the same region and merging the m adjacent points to obtain merged seed points, replacing the corresponding initial seed points, adding the merged seed points into the specific sequence until the sequence is empty, otherwise, deleting the adjacent points of which the curvature values are larger than the threshold value, and repeatedly checking the adjacent points in the region until no point can be merged, thereby finishing the growth.
Iterating the processing process of the adjacent point of the area, calculating curvature values of m adjacent points of the seed point after the first combination in the specific sequence, if the curvature values of the m adjacent points are smaller than a threshold value, considering that the m adjacent points belong to the same area and combine the m adjacent points to obtain the seed point after the second combination, replacing the corresponding seed point after the first combination, adding the seed point into the specific sequence until the sequence is empty, otherwise, deleting the adjacent points of which the curvature values are larger than the threshold value, repeatedly checking the adjacent points in the area until no point can be combined, and thus finishing the growth. And by analogy, performing iterative processing, marking all points in the point cloud cluster into different regions, and obtaining the three-dimensional point cloud cluster of the target object until all the points are marked into different regions.
Referring to fig. 2, a specific embodiment of the present invention shows a process for acquiring point cloud data by the airborne radar of the present invention.
Before the unmanned aerial vehicle flies, firstly, determining a no-fly area in a local air traffic control bureau, applying a flying airspace, planning a flight path, performing on-site reconnaissance and checking equipment accessories; after accomplishing these works, check out test set guarantees that unmanned aerial vehicle and laser radar equipment are normal, guarantees that unmanned aerial vehicle battery power is full of, carries out the on-the-spot unloaded flight test.
Data collection: in the field line patrol flight process, firstly, flight carrying laser radar with the line overlapping degree not less than two thirds is carried out according to a preset route prepared by flight, a base station collects GNSS + IMU data, and the laser radar collects LIDAR data and image data simultaneously.
Data arrangement: and transmitting the data to a base station, and performing combined solution of GNSS and IMU by the base station to draw the unmanned aerial vehicle track data in real time.
The invention takes the example of dividing the tunnel portal in the railway scene as the following test steps.
1. When point cloud data is obtained, due to the influences of equipment precision, obstacles and the like, some noise points and outliers can appear, so that the test adopts direct filtering and statistical outlier filtering to carry out secondary filtering. And finally, obtaining complete point cloud data. Fig. 3 shows a schematic diagram of the complete point cloud data.
2. And carrying out point cloud segmentation on the preprocessed three-dimensional point cloud.
2.1, carrying out point cloud segmentation on the three-dimensional point cloud after the second-stage filtering by using a region growing method, and setting a parameter region:
(1) the number of the logout points is 500-;
(2) the maximum point number of the region is 50000-200000, and 100000 is taken in the experiment;
(3) the neighborhood m is 100-500, and 120 is taken in the experiment;
(4) the normal line included angle threshold value is 1-10 degrees, and the angle is 5 degrees in the experiment;
(5) the curvature threshold is 0.10-0.20, and 0.12 is taken in the experiment.
3. Selecting a relatively flat rail below a tunnel portal as a seed point P, calculating the curvature of the point P to be Kp =0.103 according to the covariance matrix, searching for a near point, and calculating an included angle between the normal of the near point and the normal of the seed point to be about 4 degrees. The neighborhood point is added to the rail area and the seed point P is deleted.
4. And iteratively executing the steps until the point cloud with the tunnel portal as the center is segmented from the scene.
5. According to the experiment, the segmentation result of the three-dimensional point cloud is shown in fig. 4.
The experimental results of the specific examples of the present invention are as follows.
Referring to fig. 4, different colors represent different divided objects, and effective point clouds of single objects can be divided, and the number of the point clouds of single objects is shown in table 1.
| Point cloud | Zone 1 (tunnel portal) | Zone 2 (Rail) | Region 3 (mountain) |
| Counting number | 48435 | 10476 | 96768 |
TABLE 1 Point count of Single object Point clouds
The point number of the single object point cloud is slightly lost, but the loss is small, and obvious tunnel junctions, rails and mountain bodies along the railway can be seen from the attached figure 4, and a better segmentation result can be obtained by adjusting and setting the threshold parameters. In summary, the process can be used for effectively carrying out scene segmentation on the railway point cloud collected by the airborne radar.
The point cloud segmentation method for area growth provides parameter threshold values in a complex railway scene, and selects proper parameters to meet segmentation requirements according to different conditions. In the segmentation process, whether the point cloud sets belong to the same object or not is judged according to whether the included angle between the seed point normal and the neighbor point normal is within a set threshold value, the error rate is low, and the problem of excessive or incomplete segmentation is effectively solved.
As described above, only the preferred embodiments of the present invention are described, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should be considered as the protection scope of the present invention.
Claims (10)
1. A point cloud segmentation method for a railway scene based on airborne radar point cloud is characterized by comprising the following steps:
s1, determining an initial seed point, and marking the initial seed point into a specific sequence;
s2, searching a near-adjacent point, and adding the near-adjacent point into a current area if an included angle between the near-adjacent point and a normal of the seed point is smaller than a threshold value;
s3, calculating a curvature value of a neighboring point, and if the curvature value is smaller than a threshold value, adding the curvature value into the specific sequence;
s4, deleting the corresponding initial seed point;
s5, repeating the steps S1-S4 until the initial seed point in the specific sequence is empty;
s6, carrying out iterative processing on the specific sequence according to the steps S2-S5 until all points are marked into different areas;
and S7, obtaining the target object.
2. The method for segmenting the point cloud of the railway scene based on the point cloud of the airborne radar according to claim 1, wherein the step S1 specifically comprises: and finding the point with the minimum curvature value from the flattest area as the initial seed point, wherein the specific sequence is a sequence formed by the initial seed points.
3. The method for segmenting the point cloud of the railway scene based on the point cloud of the airborne radar according to claim 1, wherein the step S2 specifically comprises: searching m neighbor points which are closest to the Euclidean distance of one initial seed point, calculating the normal direction of each neighbor point and the normal direction of the corresponding initial seed point, and adding the neighbor point to the current area if the included angle between the normal of the neighbor point and the normal of the initial seed point is smaller than a normal included angle threshold value.
4. The method for segmenting the point cloud of the railway scene based on the point cloud of the airborne radar according to claim 1, wherein the step S3 specifically comprises: and calculating curvature values of the m adjacent points of the initial seed point, if the curvature values of the m adjacent points are smaller than a curvature threshold value, considering that the m adjacent points belong to the same region and combining the m adjacent points to obtain a combined seed point, and adding the combined seed point into the specific sequence.
5. The method for point cloud segmentation of railway scene based on airborne radar point cloud according to claim 4, wherein the step S6 specifically comprises: and performing iterative processing on the combined seed points in the specific sequence according to the steps S2-S5 until all the points are marked into different areas.
6. The method for point cloud segmentation of railway scene based on airborne radar point cloud according to claim 4, wherein before the step S1, the method further comprises: and acquiring point cloud data, and performing secondary filtering by adopting direct filtering and statistical outlier filtering to obtain complete point cloud data.
7. The method for segmenting a point cloud of a railway scene based on a point cloud of an airborne radar according to claim 6, wherein the acquiring the point cloud data specifically comprises: using an unmanned aerial vehicle to carry out on-site line patrol flight, carrying out laser radar-carrying flight with the line overlapping degree not less than two thirds according to a preset route, and simultaneously acquiring LIDAR data and image data by using the laser radar; and transmitting the data to a base station, and performing combined solution of GNSS and IMU by the base station to obtain point cloud data.
8. The airborne radar point cloud-based railway scene point cloud segmentation method according to claim 3, wherein the threshold value of the normal included angle is 1-10 °.
9. The method of point cloud segmentation for railway scene based on airborne radar point cloud according to claim 4, wherein the curvature threshold is 0.10-0.20.
10. The method for point cloud segmentation of railway scene based on airborne radar point cloud according to claim 8 or 9, characterized in that the threshold value of normal angle is 5 ° and the threshold value of curvature is 0.12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210133610.3A CN114494301A (en) | 2022-02-14 | 2022-02-14 | Railway scene point cloud segmentation method based on airborne radar point cloud |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210133610.3A CN114494301A (en) | 2022-02-14 | 2022-02-14 | Railway scene point cloud segmentation method based on airborne radar point cloud |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114494301A true CN114494301A (en) | 2022-05-13 |
Family
ID=81480598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210133610.3A Pending CN114494301A (en) | 2022-02-14 | 2022-02-14 | Railway scene point cloud segmentation method based on airborne radar point cloud |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114494301A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409437A (en) * | 2018-11-06 | 2019-03-01 | 安徽农业大学 | A kind of point cloud segmentation method, apparatus, computer readable storage medium and terminal |
| CN109934855A (en) * | 2018-12-28 | 2019-06-25 | 南京理工大学 | A kind of livewire work scene power components three-dimensional rebuilding method based on cloud |
| CN110992341A (en) * | 2019-12-04 | 2020-04-10 | 沈阳建筑大学 | Segmentation-based airborne LiDAR point cloud building extraction method |
| CN113298833A (en) * | 2021-05-20 | 2021-08-24 | 山东大学 | Target object point cloud characteristic line and surface extraction method and system |
| CN113325428A (en) * | 2021-04-29 | 2021-08-31 | 中国科学院沈阳自动化研究所 | Distance and course measuring method based on laser radar |
| CN113569313A (en) * | 2021-07-12 | 2021-10-29 | 北京铁科特种工程技术有限公司 | Railway fence modeling and identifying method based on laser radar secondary echo point cloud data |
-
2022
- 2022-02-14 CN CN202210133610.3A patent/CN114494301A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409437A (en) * | 2018-11-06 | 2019-03-01 | 安徽农业大学 | A kind of point cloud segmentation method, apparatus, computer readable storage medium and terminal |
| CN109934855A (en) * | 2018-12-28 | 2019-06-25 | 南京理工大学 | A kind of livewire work scene power components three-dimensional rebuilding method based on cloud |
| CN110992341A (en) * | 2019-12-04 | 2020-04-10 | 沈阳建筑大学 | Segmentation-based airborne LiDAR point cloud building extraction method |
| CN113325428A (en) * | 2021-04-29 | 2021-08-31 | 中国科学院沈阳自动化研究所 | Distance and course measuring method based on laser radar |
| CN113298833A (en) * | 2021-05-20 | 2021-08-24 | 山东大学 | Target object point cloud characteristic line and surface extraction method and system |
| CN113569313A (en) * | 2021-07-12 | 2021-10-29 | 北京铁科特种工程技术有限公司 | Railway fence modeling and identifying method based on laser radar secondary echo point cloud data |
Non-Patent Citations (1)
| Title |
|---|
| 张楠: "铁路场景下三维点云识别与分类算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109949326B (en) | Building contour line extraction method based on knapsack type three-dimensional laser point cloud data | |
| CN108109437B (en) | Unmanned aerial vehicle autonomous route extraction and generation method based on map features | |
| US10198632B2 (en) | Survey data processing device, survey data processing method, and survey data processing program | |
| CN110490415B (en) | Building progress assessment method for visual coordination of multiple unmanned aerial vehicles | |
| CN106290388A (en) | A kind of insulator breakdown automatic testing method | |
| US9384399B2 (en) | Method and system for processing image data obtained from scanning a network infrastructure | |
| CN106778749B (en) | Method for extracting border of itinerant operation area based on concentration and Delaunay triangular reconstruction | |
| CN101915570B (en) | Vanishing point based method for automatically extracting and classifying ground movement measurement image line segments | |
| CN113325868B (en) | Crop real-time identification system and method based on unmanned aerial vehicle | |
| CN108267121A (en) | The vision navigation method and system of more equipment under a kind of variable scene | |
| CN106680798A (en) | Airborne LIDAR air strips overlay region redundancy identification and elimination method | |
| CN113554595A (en) | Unmanned aerial vehicle laser radar point cloud tower head deformation detection device and method | |
| CN115690081A (en) | Tree counting method, system, storage medium, computer equipment and terminal | |
| Vick et al. | A review of linear transportation construction progress monitoring techniques | |
| CN116486289A (en) | Gas pipeline high-consequence area identification method driven by multi-source data and knowledge | |
| CN107133236A (en) | The method and apparatus of geographical position track optimizing | |
| CN116071666A (en) | Bare soil dust source monitoring method based on high-resolution satellite image | |
| Neubert et al. | Extraction of railroad objects from very high resolution helicopter-borne LiDAR and ortho-image data | |
| CN114004950A (en) | Intelligent pavement disease identification and management method based on BIM and LiDAR technology | |
| CN116740060B (en) | Method for detecting size of prefabricated part based on point cloud geometric feature extraction | |
| CN114494301A (en) | Railway scene point cloud segmentation method based on airborne radar point cloud | |
| CN112097743A (en) | Building earthquake-resistant characteristic analysis method based on unmanned aerial vehicle technology | |
| CN110148218B (en) | Method for integrally optimizing large-batch airborne LiDAR point cloud data | |
| CN110807832A (en) | Automatic power line extraction method based on ground three-dimensional scanner | |
| CN117253205A (en) | Road surface point cloud rapid extraction method based on mobile measurement system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220513 |