CN109858571B - Laser radar point cloud power line classification method based on normal distribution and clustering - Google Patents
Laser radar point cloud power line classification method based on normal distribution and clustering Download PDFInfo
- Publication number
- CN109858571B CN109858571B CN201910182611.5A CN201910182611A CN109858571B CN 109858571 B CN109858571 B CN 109858571B CN 201910182611 A CN201910182611 A CN 201910182611A CN 109858571 B CN109858571 B CN 109858571B
- Authority
- CN
- China
- Prior art keywords
- power line
- point
- cube
- data set
- point cloud
- 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
Images
Abstract
The invention discloses a laser radar point cloud power line classification method based on normal distribution and clustering, which comprises the following steps of: (1) preprocessing original point cloud data, establishing a digital terrain model, and performing power line candidate point rough extraction by using elevation filtering; (2) aiming at the power line candidate point data after the rough extraction, further optimizing and extracting the power line candidate points in the three-dimensional space based on a normal distribution transformation algorithm; (3) and aiming at the power line candidate point data after optimized extraction, the accurate extraction of the power line points is realized by using a mean value clustering algorithm. The method can realize power line classification in the laser radar point cloud data of various complex environments such as urban forest regions and the like, provides accurate power line extraction results, greatly improves the point cloud data classification efficiency, provides a new thought for various point cloud data classification, and provides accurate and comprehensive analysis data for power line inspection and other work.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a laser radar point cloud power line classification method based on normal distribution and clustering.
Background
The laser radar technology is a novel and efficient space detection means, a large amount of point cloud data with accurate three-dimensional space coordinates of a target scene can be rapidly acquired in a short time, but compared with the progress of hardware performance and indexes of a laser radar system, software processing of the laser radar point cloud data is still in an initial stage, and the main problem in the current laser radar point cloud data processing field is how to effectively utilize the mass point cloud data acquired by the laser radar hardware system. Meanwhile, with the rapid development of economy in various fields of China, the demand of various industries on electric power is increased very rapidly. In the face of a large-scale power network and a corresponding power communication network, in the practical problems of new power grid site planning, power line optimization, power grid safety management and maintenance, power optical fiber cable network safety management and maintenance and the like, an efficient and reliable measurement technology is urgently needed to meet the economic and technical requirements.
The data along the power grid collected and processed by the laser radar measurement technology can recover the actual geometric information of the electric wire and the electric power optical cable, further automatically measure the distance between the electric wire and the ground and the distance between adjacent electric wires, acquire sag, span and the like of the electric wire and the electric power communication optical cable through related calculation, and realize the derivation of direct safety parameters such as the tension or the pulling force of the electric wire and the electric power optical cable through the geometric parameters. Therefore, by utilizing the laser radar point cloud data, the automatic extraction method of the power line can quickly acquire related spatial data and provide support for refinement, scientification, high efficiency and the like of power grid design and management.
Disclosure of Invention
In order to solve the technical problems, the invention provides the laser radar point cloud power line classification method based on normal distribution and clustering, which has high classification efficiency and high extraction precision.
The technical scheme for solving the problems is as follows: a laser radar point cloud power line classification method based on normal distribution and clustering comprises the following steps:
(1) preprocessing original point cloud data, establishing a digital terrain model, and performing power line candidate point rough extraction by using elevation filtering;
(2) aiming at the power line candidate point data after the rough extraction, further optimizing and extracting the power line candidate points in the three-dimensional space based on a normal distribution transformation algorithm;
(3) and aiming at the power line candidate point data after optimized extraction, the accurate extraction of the power line points is realized by using a mean value clustering algorithm.
The laser radar point cloud power line classification extraction method based on normal distribution and clustering comprises the specific steps of the step (1):
1-1) carrying out point cloud data preprocessing based on original point cloud data and a filtering mechanism of obvious non-power line points;
1-2) describing a scene according to the quality of original point cloud data, and designing a ground seed point distance of 0.5 m;
1-3) carrying out block processing on the point cloud data to obtain a plurality of small areas, and selecting a ground seed point in each small area;
1-4) carrying out ground fitting by using the obtained seed points to obtain a digital terrain model which is recorded as DTM;
1-5) subtracting the DTM from the original point cloud elevation to obtain point cloud relative height information;
1-6) extracting points with relative heights of 4m and more than 4m as a power line candidate point data set one.
The laser radar point cloud power line classification extraction method based on normal distribution and clustering comprises the specific steps of the step (2):
2-1) the coordinates of all power line candidate points in the first power line candidate point data set are x, y and z, the difference values of the maximum value and the minimum value of the three axes of the x, y and z of the power line candidate point coordinates are respectively calculated, and three corresponding minimum integers not less than the difference values are respectively taken as Ax、Ay、Az；
2-2) construction of size Ax*Ay*AzThe minimum value in the X-axis direction of the cube is the minimum value of the candidate point coordinate X, and the maximum value in the X-axis direction of the cube is the minimum integer not less than the maximum value of the point coordinate X; the minimum value in the cube Y-axis direction is the minimum value of the point coordinate Y, and the maximum value in the cube Y-axis direction is the minimum integer not less than the maximum value of the point coordinate Y; the minimum value of the cube Z-axis direction is the minimum value of the point coordinate Z, and the maximum value of the cube Z-axis direction is the minimum integer not less than the maximum value of the point coordinate Z;
2-3) dividing the constructed cube into n cube cells of 1 m;
2-4) distributing the power line candidate points into cubic cells meeting the coordinate position according to the coordinate values of the three-dimensional points;
2-5) counting the number of points in each cubic unit, and if the number of the points is less than 3, removing the cubic unit and the points in the cubic unit from a first data set to obtain a first optimized and updated candidate power line point data set and n' updated cubic units;
2-6) calculating the covariance matrix of the point coordinates in each cubic unit from the data set I obtained in the previous step and n' cubic units thereof, wherein the calculation formula is as follows:
in the formula: cov denotes a covariance calculation operator; x, Y and Z are coordinate values of X axis, Y axis and Z axis of the point set in each cubic unit;
then the coordinate values of the X axis, the Y axis and the Z axis of the point set in each cubic unit are the average value; m is the number of points in the corresponding cube; r is a three-dimensional covariance matrix of the point set in the corresponding cube unit;
2-7) calculating the covariance matrix R of the cubic unit according to 2-6), and calculating three eigenvalues lambda according to a conventional matrix algorithm1、λ2And λ3And setting:
λ1≤λ2≤λ3 (5)；
2-8) setting the threshold t to 0.02, if λ2/λ3If the t is less than or equal to t, marking the cubic unit as a linear cubic unit, otherwise, marking the cubic unit as a non-linear cubic unit;
2-9) extracting the linear cubic cell and the point data therein as a power line candidate point data set two.
The laser radar point cloud power line classification extraction method based on normal distribution and clustering comprises the specific steps of the step (3):
3-1) storing a feature vector corresponding to the maximum feature value in each cube unit in the data set II as sample data of a mean value clustering algorithm, and recording the sample data as a data set III;
3-2) visually selecting k characteristic vectors belonging to the cubic units of the candidate power line points in the data set III as seed vectors;
3-3) traversing the data set III, and calculating the included angle alpha between each feature vector and k seed vectors respectivelypq(ii) a Wherein p represents a feature vector corresponding to the maximum feature value of the p-th cubic unit in the data set III; q represents the qth feature vector in the k seed vectors; alpha is alphapqRepresenting the included angle between the p-th characteristic vector and the q-th seed vector in the data set III;
3-4) setting an included angle threshold value delta to be 0.5 degrees, and judging an included angle alpha between each eigenvector and k seed vectors in the data set IIIpqWith respect to the magnitude of the angle threshold Δ, if αpq<Δ, then αpqThe characteristic vector of the p-th cube unit in the corresponding data set III belongs to the q-th seed vector group in the k seed vectors, and k seed vector groups are obtained in total if alpha ispq>Delta, the p-th cube unit feature vector does not belong to any seed vector group;
3-5) calculating a normal vector of each feature vector group, and taking the k normal vectors as new seed vectors;
3-6) repeating the steps 3-3) -3-5) until the normal vector is consistent with the seed vector, and ending the circulation;
3-7) removing the cube unit where the feature vector which does not meet the included angle threshold value is located and the point data in the cube unit, wherein the remaining point data in the third data set is the final power line point, and the point data in the cube unit where the k feature vector groups are located belongs to k power lines in different directions.
The invention has the beneficial effects that: the invention provides a method for power lines in laser radar point cloud based on normal distribution and clustering. The method can realize power line classification in laser radar point cloud data of various complex environments such as urban forest regions and the like, provides accurate power line extraction results, greatly improves point cloud data classification efficiency, provides a new thought for various point cloud data classification, and provides accurate and comprehensive analysis data for work such as power line inspection and the like.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a schematic diagram of the coincidence of the normal vector and the seed vector in step (3) of the present invention.
Detailed Description
The invention is further described below with reference to the figures and examples.
As shown in fig. 1, a method for classifying laser radar point cloud power lines based on normal distribution and clustering includes the following steps:
(1) the method comprises the steps of preprocessing original point cloud data, establishing a digital terrain model, and performing power line candidate point rough extraction by using elevation filtering. The specific steps of the step (1) are as follows:
1-1) carrying out point cloud data preprocessing based on original point cloud data and a conventional and traditional filtering mechanism of remarkable non-power line points (noise points, error and leakage points and the like);
1-2) describing a scene according to the quality of original point cloud data, and designing a ground seed point distance of 0.5 m;
1-3) carrying out block processing on the point cloud data to obtain a plurality of small areas, and selecting a ground seed point in each small area;
1-4) carrying out ground fitting by using the obtained seed points to obtain a digital terrain model which is recorded as DTM;
1-5) subtracting the DTM from the original point cloud elevation to obtain point cloud relative height information;
1-6) extracting points with relative heights of 4m and more than 4m as a power line candidate point data set one.
(2) And aiming at the power line candidate point data after the rough extraction, further optimizing and extracting the power line candidate points in the three-dimensional space based on a normal distribution transformation algorithm. The specific steps of the step (2) are as follows:
2-1) the coordinates of all power line candidate points in the first power line candidate point data set are x, y and z, the difference values of the maximum value and the minimum value of the three axes of the x, y and z of the power line candidate point coordinates are respectively calculated, and three corresponding minimum integers not less than the difference values are respectively taken as Ax、Ay、Az；
2-2) construction of size Ax*Ay*AzThe minimum value in the X-axis direction of the cube is the minimum value of the candidate point coordinate X, and the maximum value in the X-axis direction of the cube is the minimum integer not less than the maximum value of the point coordinate X; the minimum value in the cube Y-axis direction is the minimum value of the point coordinate Y, and the maximum value in the cube Y-axis direction is the minimum integer not less than the maximum value of the point coordinate Y; the minimum value of the cube Z-axis direction is the minimum value of the point coordinate Z, and the maximum value of the cube Z-axis direction is the minimum integer not less than the maximum value of the point coordinate Z;
2-3) dividing the constructed cube into n cube cells of 1 m;
2-4) distributing the power line candidate points into cubic cells meeting the coordinate position according to the coordinate values of the three-dimensional points;
2-5) counting the number of points in each cubic unit, and if the number of the points is less than 3, removing the cubic unit and the points in the cubic unit from a first data set to obtain a first optimized and updated candidate power line point data set and n' updated cubic units;
2-6) calculating the covariance matrix of the point coordinates in each cubic unit from the data set I obtained in the previous step and n' cubic units thereof, wherein the calculation formula is as follows:
in the formula: cov denotes a covariance calculation operator; x, Y and Z are coordinate values of X axis, Y axis and Z axis of the point set in each cubic unit;
then the coordinate values of the X axis, the Y axis and the Z axis of the point set in each cubic unit are the average value; m is the number of points in the corresponding cube; r is a three-dimensional covariance matrix of the point set in the corresponding cube unit;
2-7) calculating the covariance matrix R of the cubic unit according to 2-6), and calculating three eigenvalues lambda according to a conventional matrix algorithm1、λ2And λ3And setting:
λ1≤λ2≤λ3 (5)；
2-8) setting the threshold t to 0.02, if λ2/λ3If the t is less than or equal to t, marking the cubic unit as a linear cubic unit, otherwise, marking the cubic unit as a non-linear cubic unit;
2-9) extracting the linear cubic cell and the point data therein as a power line candidate point data set two.
(3) And aiming at the power line candidate point data after optimized extraction, the accurate extraction of the power line points is realized by using a mean value clustering algorithm. The specific steps of the step (3) are as follows:
3-1) storing a feature vector corresponding to the maximum feature value in each cube unit in the data set II as sample data of a mean value clustering algorithm, and recording the sample data as a data set III;
3-2) visually selecting k characteristic vectors belonging to the cubic units of the candidate power line points in the data set III as seed vectors;
3-3) traversing the data set III, and calculating the included angle alpha between each feature vector and k seed vectors respectivelypq(ii) a Wherein p represents a feature vector corresponding to the maximum feature value of the p-th cubic unit in the data set III; q represents the qth feature vector in the k seed vectors; alpha is alphapqRepresenting the included angle between the p-th characteristic vector and the q-th seed vector in the data set III;
3-4) setting an included angle threshold value delta to be 0.5 degrees, and judging an included angle alpha between each eigenvector and k seed vectors in the data set IIIpqWith respect to the magnitude of the angle threshold Δ, if αpq<Δ, then αpqThe characteristic vector of the p-th cube unit in the corresponding data set III belongs to the q-th seed vector group in the k seed vectors, and k seed vector groups are obtained in total if alpha ispq>Delta, the p-th cube unit feature vector does not belong to any seed vector group;
3-5) calculating a normal vector of each feature vector group, and taking the k normal vectors as new seed vectors;
3-6) repeating the steps 3-3) -3-5) until the normal vector is consistent with the seed vector, and ending the loop (as shown in fig. 2, until all point values are distributed to a certain cluster in the k normal vector groups);
3-7) removing the cube unit where the feature vector which does not meet the included angle threshold value is located and the point data in the cube unit, wherein the remaining point data in the third data set is the final power line point, and the point data in the cube unit where the k feature vector groups are located belongs to k power lines in different directions.
Claims (3)
1. A laser radar point cloud power line classification method based on normal distribution and clustering comprises the following steps:
(1) preprocessing original point cloud data, establishing a digital terrain model, and performing power line candidate point rough extraction by using elevation filtering to obtain a first power line candidate point data set;
(2) aiming at the power line candidate point data after the rough extraction, further optimizing and extracting the power line candidate points in the three-dimensional space based on a normal distribution transformation algorithm to obtain a power line candidate point data set II;
(3) aiming at the power line candidate point data after optimized extraction, the accurate extraction of the power line points is realized by using a mean value clustering algorithm;
the specific steps of the step (3) are as follows:
3-1) storing a feature vector corresponding to the maximum feature value in each cube unit in the data set II as sample data of a mean value clustering algorithm, and recording the sample data as a data set III;
3-2) visually selecting k characteristic vectors belonging to the cubic units of the candidate power line points in the data set III as seed vectors;
3-3) traversing the data set III, and calculating the included angle alpha between each feature vector and k seed vectors respectivelypq(ii) a Wherein p represents a feature vector corresponding to the maximum feature value of the p-th cubic unit in the data set III; q represents the qth feature vector in the k seed vectors; alpha is alphapqRepresenting the included angle between the p-th characteristic vector and the q-th seed vector in the data set III;
3-4) setting an included angle threshold value delta to be 0.5 degrees, and judging an included angle alpha between each eigenvector and k seed vectors in the data set IIIpqWith respect to the magnitude of the angle threshold Δ, if αpq<Δ, then αpqThe characteristic vector of the p-th cube unit in the corresponding data set III belongs to the q-th seed vector group in the k seed vectors, and k seed vector groups are obtained in total if alpha ispq>Delta, the p-th cube unit feature vector does not belong to any seed vector group;
3-5) calculating a normal vector of each feature vector group, and taking the k normal vectors as new seed vectors;
3-6) repeating the steps 3-3) -3-5) until the normal vector is consistent with the seed vector, and ending the circulation;
3-7) removing the cube unit where the feature vector which does not meet the included angle threshold value is located and the point data in the cube unit, wherein the remaining point data in the third data set is the final power line point, and the point data in the cube unit where the k feature vector groups are located belongs to k power lines in different directions.
2. The normal distribution and clustering-based laser radar point cloud power line classification method according to claim 1, wherein: the specific steps of the step (1) are as follows:
1-1) carrying out point cloud data preprocessing based on original point cloud data and a filtering mechanism of obvious non-power line points;
1-2) describing a scene according to the quality of original point cloud data, and designing a ground seed point distance of 0.5 m;
1-3) carrying out block processing on the point cloud data to obtain a plurality of small areas, and selecting a ground seed point in each small area;
1-4) carrying out ground fitting by using the obtained seed points to obtain a digital terrain model which is recorded as DTM;
1-5) subtracting the DTM from the original point cloud elevation to obtain point cloud relative height information;
1-6) extracting points with the relative height of more than 4m as a power line candidate point data set I.
3. The normal distribution and clustering-based laser radar point cloud power line classification method according to claim 2, wherein: the specific steps of the step (2) are as follows:
2-1) the coordinates of all power line candidate points in the first power line candidate point data set are x, y and z, the difference values of the maximum value and the minimum value of the three axes of the x, y and z of the power line candidate point coordinates are respectively calculated, and three corresponding minimum integers not less than the difference values are respectively taken as Ax、Ay、Az；
2-2) construction of size Ax*Ay*AzThe minimum value in the X-axis direction of the cube is the minimum value of the candidate point coordinate X, and the maximum value in the X-axis direction of the cube is the minimum integer not less than the maximum value of the point coordinate X; the minimum value in the cube Y-axis direction is the minimum value of the point coordinate Y, and the maximum value in the cube Y-axis direction is the minimum integer not less than the maximum value of the point coordinate Y; the minimum value in the cube Z-axis direction is the minimum value of the point coordinate Z, and the maximum value in the cube Z-axis direction is not less thanThe minimum integer of the maximum value of the point coordinate z;
2-3) dividing the constructed cube into n cube cells of 1 m;
2-4) distributing the power line candidate points into cubic cells meeting the coordinate position according to the coordinate values of the three-dimensional points;
2-5) counting the number of points in each cubic unit, and if the number of the points is less than 3, removing the cubic unit and the points in the cubic unit from a first data set to obtain a first optimized and updated candidate power line point data set and n' updated cubic units;
2-6) calculating the covariance matrix of the point coordinates in each cubic unit from the data set I obtained in the previous step and n' cubic units thereof, wherein the calculation formula is as follows:
in the formula: cov denotes a covariance calculation operator; x, Y and Z are coordinate values of X axis, Y axis and Z axis of the point set in each cubic unit;
then the coordinate values of the X axis, the Y axis and the Z axis of the point set in each cubic unit are the average value; m is the number of points in the corresponding cube; r is the three-dimensional covariance matrix of the point set in the corresponding cube unitArraying;
2-7) calculating the covariance matrix R of the cubic unit according to 2-6), and calculating three eigenvalues lambda according to a conventional matrix algorithm1、λ2And λ3And setting:
λ1≤λ2≤λ3 (5)；
2-8) setting the threshold t to 0.02, if λ2/λ3If the t is less than or equal to t, marking the cubic unit as a linear cubic unit, otherwise, marking the cubic unit as a non-linear cubic unit;
2-9) extracting the linear cubic cell and the point data therein as a power line candidate point data set two.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910182611.5A CN109858571B (en) | 2019-03-12 | 2019-03-12 | Laser radar point cloud power line classification method based on normal distribution and clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910182611.5A CN109858571B (en) | 2019-03-12 | 2019-03-12 | Laser radar point cloud power line classification method based on normal distribution and clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109858571A CN109858571A (en) | 2019-06-07 |
| CN109858571B true CN109858571B (en) | 2021-07-30 |
Family
ID=66900531
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910182611.5A Active CN109858571B (en) | 2019-03-12 | 2019-03-12 | Laser radar point cloud power line classification method based on normal distribution and clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109858571B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111814666B (en) * | 2020-07-07 | 2021-09-24 | 华中农业大学 | Single tree parameter extraction method, system, medium and equipment under complex forest stand |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102590823A (en) * | 2012-01-06 | 2012-07-18 | 中国测绘科学研究院 | Rapid extraction and reconstruction method for data power line of airborne LIDAR |
| US9052721B1 (en) * | 2012-08-28 | 2015-06-09 | Google Inc. | Method for correcting alignment of vehicle mounted laser scans with an elevation map for obstacle detection |
| CN107292335A (en) * | 2017-06-06 | 2017-10-24 | 云南电网有限责任公司信息中心 | A kind of transmission line of electricity cloud data automatic classification method based on Random Forest model |
-
2019
- 2019-03-12 CN CN201910182611.5A patent/CN109858571B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102590823A (en) * | 2012-01-06 | 2012-07-18 | 中国测绘科学研究院 | Rapid extraction and reconstruction method for data power line of airborne LIDAR |
| US9052721B1 (en) * | 2012-08-28 | 2015-06-09 | Google Inc. | Method for correcting alignment of vehicle mounted laser scans with an elevation map for obstacle detection |
| CN107292335A (en) * | 2017-06-06 | 2017-10-24 | 云南电网有限责任公司信息中心 | A kind of transmission line of electricity cloud data automatic classification method based on Random Forest model |
Non-Patent Citations (4)
| Title |
|---|
| Fully-Automated Power Line Extraction from Airborne Laser Scanning Point Clouds in Forest Areas;lingli zhu et al.;《remote sensing》;20141113;第6卷(第11期);第11267-11282页 * |
| Supervised Classification of Power Lines from Airborne LiDAR Data in Urban Areas;Wang, YJ et al.;《REMOTE SENSING》;20170728;第9卷(第8期) * |
| Voxel-Based Extraction of Transmission Lines From Airborne LiDAR Point Cloud Data;Juntao Yang et al.;《IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing》;20180926;第11卷(第10期);第3892-3904页 * |
| 贾魁 等.多尺度下车载激光点云数据中电力线的提取.《 首都师范大学学报(自然科学版)》.2018,第39卷(第1期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109858571A (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829199B (en) | Power line rapid layered extraction method based on LiDAR point cloud | |
| CN110675441B (en) | Laser point cloud-based power transmission line ground wire modeling extraction method | |
| CN106599915B (en) | A kind of vehicle-mounted laser point cloud classifications method | |
| CN113592822B (en) | Insulator defect positioning method for electric power inspection image | |
| CN112883878B (en) | Automatic point cloud classification method under transformer substation scene based on three-dimensional grid | |
| CN109858571B (en) | Laser radar point cloud power line classification method based on normal distribution and clustering | |
| CN115825984A (en) | Power transmission line inspection system and inspection method based on laser point cloud | |
| CN115619972A (en) | Road marking abrasion identification and evaluation method based on point cloud data | |
| CN112561989B (en) | Recognition method for hoisting object in construction scene | |
| CN112529044B (en) | Method for extracting and classifying railway contact network based on vehicle-mounted LiDAR | |
| CN109948682B (en) | Laser radar point cloud power line classification method based on normal random sampling distribution | |
| Li | Typical trajectory extraction method for ships based on AIS data and trajectory clustering | |
| CN117371949A (en) | Three-dimensional visual model-based power transmission line construction safety monitoring method and system | |
| CN115757369A (en) | Automatic inspection method and system for laser point cloud data | |
| Guo et al. | Research on point cloud power line segmentation and fitting algorithm | |
| CN115170950A (en) | Outdoor scene building extraction method based on multi-feature constraint | |
| CN113205543A (en) | Laser radar point cloud trunk extraction method based on machine learning | |
| CN113192172B (en) | Airborne LiDAR ground point cloud simplification method | |
| Wang et al. | Banana pseudostem width detection based on kinect V2 depth sensor | |
| CN117132915B (en) | Power transmission line tree obstacle hidden danger analysis method based on automatic classification of point cloud | |
| Li et al. | Optimization of Positioning Error and Time for UWB and Virtual-Real Scenes Similarity Measure Based on Error Model | |
| CN112882047B (en) | Laser point cloud based automatic acquisition method and system for power line inter-phase distance | |
| Shi et al. | Application of artificial intelligence technology in early warning of power infrastructure construction | |
| CN117437630A (en) | Overhead line point cloud detection method based on catenary surface RANSAC segmentation algorithm | |
| Shihua et al. | An Individual Tree Segmentation Algorithm Without Prior-Knowledge Based on Airborne Lidar Data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |