CN117689965A - Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm - Google Patents
Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm Download PDFInfo
- Publication number
- CN117689965A CN117689965A CN202410153806.8A CN202410153806A CN117689965A CN 117689965 A CN117689965 A CN 117689965A CN 202410153806 A CN202410153806 A CN 202410153806A CN 117689965 A CN117689965 A CN 117689965A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- rock mass
- structural surface
- cloud data
- 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
- 239000011435 rock Substances 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 39
- 239000013598 vector Substances 0.000 claims abstract description 34
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 4
- 206010057175 Mass conditions Diseases 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 238000004441 surface measurement Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides a rapid identification method of a laser point cloud rock mass structure based on a RANSAC algorithm, which comprises the following steps: scanning the rock mass structural plane by using a three-dimensional laser scanner to obtain point cloud data of the rock mass structural plane; step 2: preprocessing the point cloud data to obtain preprocessed point cloud data, establishing a data index for the preprocessed point cloud, and solving a solution vector for each point by using a nearest neighbor method to obtain the point cloud containing a normal vector; step 3: extracting a structural surface point cloud from the point cloud containing the normal vector by using a RANSAC algorithm, and acquiring an optimized structural surface point set; step 4: and identifying the structural surface according to the optimized structural surface point set by using a region growing algorithm, and calculating the tendency and the inclination angle of the structural surface. According to the jointed rock mass model and the jointed inclination angle analysis generated by the invention, more accurate and more real rock mass conditions can be obtained, and more reliable basis is provided for engineering sites.
Description
Technical Field
The invention relates to the field of geotechnical engineering, in particular to a rapid identification method for a laser point cloud rock mass structure based on a RANSAC algorithm.
Background
The rock mass is a geologic body with discontinuity, heterogeneity and anisotropy, which is composed of various rocks containing weak structural surfaces in a certain engineering range. Rock mass is formed during a lengthy geological history, with a certain structure and construction. Rock mass is composed of various rocks and is subject to various internal and external geological effects such as structural changes, weathering and unloading during its formation and is modified, so that the rock mass is often cut by various structural planes, making the rock mass a multi-fracture discontinuous medium.
The three-dimensional laser scanning technology is a technology for rapidly reconstructing a three-dimensional model of a measured object by recording information such as three-dimensional coordinates, reflectivity, texture and the like of a large number of dense points on the surface of the measured object by utilizing the principle of laser ranging. Since three-dimensional laser scanning systems can densely acquire a large number of data points of a target object, three-dimensional laser scanning techniques are also referred to as revolutionary technological breakthroughs from single point measurement to surface measurement, as opposed to conventional single point measurement. The three-dimensional laser scanning is used for scanning the rock mass surface to obtain point cloud data of the rock mass surface, a rock mass three-dimensional grid model is reconstructed according to the point cloud data, and the rock mass surface structural plane is accurately identified. The plane fitting technology of the point cloud mainly comprises a least square fitting method and a random sampling consensus method (RANSAC), wherein the traditional least square fitting method only considers errors in the observed vector, ignores errors in a coefficient matrix, has poor precision of a fitting plane and does not have robustness. Random sample consensus (RANSAC) is an algorithm that determines an initial model with random sampling and iteratively seeks optimal parameter estimates multiple times based on distance thresholds, which can fit a plane model of a specified law in the case where a large number of noise points are contained in the data.
The existing rock mass structural plane identification technology is mainly based on image identification, and is characterized in that an excavated plane image is acquired, and the rock mass structural plane is segmented and identified by using a pre-trained rock mass structural plane identification model. However, the method has extremely high requirements on the quality of the acquired image, can not accurately eliminate redundant noise points, and has low structural plane recognition accuracy.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides the rapid identification method of the laser point cloud rock mass structure based on the RANSAC algorithm, which can be used for identification under construction conditions, has higher identification accuracy and provides more accurate support for on-site geological analysis.
A rapid identification method of a laser point cloud rock mass structure based on a RANSAC algorithm comprises the following steps:
step 1: scanning the rock mass structural plane by using a three-dimensional laser scanner to obtain point cloud data of the rock mass structural plane;
step 2: preprocessing the point cloud data to obtain preprocessed point cloud data, establishing a point cloud data index for the preprocessed point cloud by adopting a kd-tree method, and solving a solution vector for each point cloud by using a nearest neighbor method to obtain the point cloud containing the normal vector;
the step of preprocessing comprises the following steps:
step 2.1: extracting the ROI region of the point cloud of the original rock mass structural surface by using CloudCompare, and removing interferents including underground pipelines and vehicles;
step 2.2: traversing the point clouds and calculating the average distance between each point cloud and n neighborhood points of the point cloud; then, calculating standard deviation of average distance between the point cloud and the neighborhood pointσAnd mean valueμDistance thresholdd max The calculation formula is thatd max =μ+α×σWhereinαIs a proportionality coefficient, i.e. a domain pointIs the number of (3); finally, traversing the point cloud again and eliminating the point cloud with the average distance larger thand max Is a point of (2).
Step 3: extracting structural surface point clouds from point cloud data containing normal vectors by adopting a RANSAC method, establishing a structural surface point cloud set, optimizing parameters of the structural surface point clouds by adopting a nonlinear model method for the point clouds in the structural surface point cloud set, and obtaining an optimized structural surface point cloud data set;
step 4: and identifying the structural surface according to the optimized structural surface point set by using a region growing algorithm, and calculating the tendency and the inclination angle of the structural surface.
Preferably, the step of establishing the point cloud data index by using the kd-Tree method in the step 2 includes:
step 2.3: sorting the preprocessed point cloud on three coordinate axes, respectively calculating standard deviation of preprocessed point cloud data on three dimensions, sorting the preprocessed point cloud data according to the standard deviation, and taking the dimension with the largest standard deviation as a first sorting coordinate axis, wherein the sequence of the standard deviation is the replacement sequence of the sorting coordinate axis;
step 2.4: calculating the median of all the point cloud data on three sequencing coordinate axes, putting the median on a current dividing node to serve as a root node of a binary tree, dividing point clouds smaller than or equal to the median into a left subtree of the binary tree, and dividing point clouds larger than the median into a right subtree of the binary tree;
step 2.5: after the sub-division is completed, the sorting coordinate axes are replaced according to the sequence in the step 2.3;
step 2.6: repeating the steps 2.4 to 2.5 until the establishment of the kd-tree index of the point cloud data is completed.
Preferably, step 3 is specifically as follows:
step 3.1: randomly extracting 4 non-collinear sample data from the point cloud data containing normal vectors, calculating a transformation matrix H, marking the transformation matrix H as a model M, and calculating a matrix H calculation formula as followsWherein (x, y) represents the target image corner position, (x ', y') is the scene image corner position, sIs a scale parameter;
step 3.2: calculating projection errors of all point cloud data containing normal vectors and a model M, and if the projection errors are smaller than a threshold value, adding the projection errors into an inner point set Q, wherein the threshold value is determined according to the size of data noise;
step 3.3: when the number of elements in the inner point set Q is greater than the optimal inner point set Q-best, then update Q-best=q, and update the maximum number of iterations at the same timeK,WhereinpIn order for the confidence level to be high,wthe proportion of the elements contained in the inner point set Q to the elements in all the point cloud data containing normal vectors; step 3.4: if the iteration number is greater than K, exiting; otherwise the iteration number is increased by 1 and steps 3.1 to 3.3 are repeated.
Further preferably, the scale parameter is the standard deviation of the point cloud data in step 3.1.
Further preferably, the threshold value in step 3.2 is taken to be 0.01.
Further preferably, the confidence level is set in step 3.3pTake 0.995.
Preferably, step 4 is specifically as follows:
step 4.1: randomly selecting one point in the point cloud area as an initial point;
step 4.2: calculating the included angle of the normal vector of the initial point and the included angles of the normal vectors of all points in the neighborhood around the initial point;
step 4.3: if the absolute value of the difference between the points in the vicinity of the initial point and the normal vector angle of the initial point is smaller than the flatness constraint thresholdεMerging the point into a point cloud area where the initial point is located;
step 4.4: searching a new initial point in the points merged into the point cloud area until no new point is merged into the point cloud area, so as to obtain a merging set of all the point cloud areas;
step 4.5: and fitting all points in each merging set to obtain the normal line of the plane of the structural surface, identifying the structural surface and calculating the inclination and the inclination angle of the structural surface.
Further preferably, in step 4.3εTake the value of0.1. The beneficial effects of the invention are as follows:
the invention provides a rapid identification method of a laser point cloud rock mass structure based on a RANSAC algorithm, which comprises the steps of scanning a rock mass structure surface by using a three-dimensional laser scanner, obtaining point cloud data of the rock mass structure surface, preprocessing the obtained point cloud data, obtaining preprocessed point cloud data, establishing a data index for the preprocessed point cloud, solving a solution vector for each point by using a nearest neighbor method, obtaining point cloud containing a normal vector, extracting the structural surface point cloud from the point cloud containing the normal vector by using the RANSAC algorithm, obtaining an optimized structural surface point set, identifying the structural surface according to the optimized structural surface point set by using a region growing algorithm, and calculating the trend and the dip angle of the structural surface.
According to the jointed rock mass model and the jointed inclination angle analysis generated by the invention, redundant noise points can be eliminated more accurately, more accurate and more real rock mass conditions are obtained, and the purposes of improving the recognition precision and recognition accuracy of the structural surface are achieved. The rock mass structural plane identification method can be used for identification under construction conditions, does not require on-site environmental quality, and provides accurate support for on-site geological analysis.
Drawings
Fig. 1 is a flow chart of a rapid identification method of a laser point cloud rock mass structure based on a RANSAC algorithm.
Detailed Description
In order that the invention may be understood, a full and complete detailed description of the invention is provided below, taken in conjunction with the accompanying drawings.
As shown in fig. 1, a rapid identification method for a laser point cloud rock mass structure based on a RANSAC algorithm includes:
step 1: and scanning the rock mass structural surface by using a three-dimensional laser scanner to obtain point cloud data of the rock mass structural surface.
Step 2: preprocessing the point cloud data to obtain preprocessed point cloud data, establishing a point cloud data index for the preprocessed point cloud by adopting a kd-tree method, and solving a solution vector for each point cloud by using a nearest neighbor method to obtain the point cloud containing the normal vector.
Step 2.1: and (3) extracting the ROI region of the point cloud of the original rock mass structural surface by using CloudCompare, and removing the interferents including underground pipelines and vehicles.
Step 2.2: traversing the point clouds and calculating the average distance between each point cloud and n neighborhood points of the point cloud; then, calculating standard deviation of average distance between the point cloud and the neighborhood pointσAnd mean valueμDistance thresholdd max The calculation formula is thatd max =μ+α×σWhereinσIs a scaling factor, i.e., the number of domain points; finally, traversing the point cloud again and eliminating the point cloud with the average distance larger thand max Is a point of (2).
Step 2.3: sorting the preprocessed point cloud on three coordinate axes, respectively calculating standard deviation of preprocessed point cloud data on three dimensions, sorting the preprocessed point cloud data according to the standard deviation, and taking the dimension with the largest standard deviation as a first sorting coordinate axis, wherein the sequence of the standard deviation is the replacement sequence of the sorting coordinate axis;
step 2.4: calculating the median of all the point cloud data on three sequencing coordinate axes, putting the median on a current dividing node to serve as a root node of a binary tree, dividing point clouds smaller than or equal to the median into a left subtree of the binary tree, and dividing point clouds larger than the median into a right subtree of the binary tree;
step 2.5: after the sub-division is completed, the sorting coordinate axes are replaced according to the sequence in the step 2.3;
step 2.6: repeating the steps 2.4 to 2.5 until the establishment of the kd-tree index of the point cloud data is completed.
Step 3: extracting structural surface point clouds from point cloud data containing normal vectors by adopting a RANSAC method, establishing a structural surface point cloud set, optimizing parameters of the structural surface point clouds by adopting a nonlinear model method for the point clouds in the structural surface point cloud set, and obtaining an optimized structural surface point cloud data set.
Step 3.1: randomly extracting 4 non-collinear sample data from the point cloud data containing the normal vector, calculating a transformation matrix H, marking the transformation matrix H as a model M, and calculating a matrix H calculation formula as followsWherein (x, y) represents the position of the corner of the target image, (x ', y') is the position of the corner of the scene image, s is a scale parameter, and s is generally taken as the standard deviation of the point cloud data.
Step 3.2: and calculating projection errors of all point cloud data containing normal vectors and the model M, and adding the projection errors into the interior point set Q if the projection errors are smaller than a threshold value, wherein the threshold value is determined according to the size of data noise and is generally taken as 0.01.
Step 3.3: when the number of elements in the inner point set Q is greater than the optimal inner point set Q-best, then the Q-best=q is updated, and the maximum iteration number is updated at the same timeK,WhereinpFor confidence, 0.995 is typically taken.wThe elements contained in the internal point set Q are the proportion of the elements in all the point cloud data containing normal vectors.
Step 3.4: if the number of iterations is greater thanKExiting; otherwise the iteration number is increased by 1 and steps 3.1 to 3.3 are repeated.
Step 4: and identifying the structural surface according to the optimized structural surface point set by using a region growing algorithm, and calculating the tendency and the inclination angle of the structural surface.
Step 4.1: randomly selecting one point in the point cloud area as an initial point;
step 4.2: calculating the included angle of the normal vector of the initial point and the included angles of the normal vectors of all points in the neighborhood around the initial point;
step 4.3: if the absolute value of the difference between the points in the vicinity of the initial point and the normal vector angle of the initial point is smaller than the flatness constraint thresholdεMerging the point into a point cloud area where the initial point is located,εthe value is generally 0.1.
Step 4.4: searching a new initial point in the points merged into the point cloud area until no new point is merged into the point cloud area, so as to obtain a merging set of all the point cloud areas;
step 4.5: and fitting all points in each merging set to obtain the normal line of the plane of the structural surface, identifying the structural surface and calculating the inclination and the inclination angle of the structural surface.
Claims (8)
1. The quick identification method of the laser point cloud rock mass structure based on the RANSAC algorithm is characterized by comprising the following steps of:
step 1: scanning the rock mass structural plane by using a three-dimensional laser scanner to obtain point cloud data of the rock mass structural plane;
step 2: preprocessing the point cloud data to obtain preprocessed point cloud data, establishing a point cloud data index for the preprocessed point cloud by adopting a kd-tree method, and solving a solution vector for each point cloud by using a nearest neighbor method to obtain the point cloud containing the normal vector;
the step of preprocessing comprises the following steps:
step 2.1: extracting the ROI region of the point cloud of the original rock mass structural surface by using CloudCompare, and removing interferents including underground pipelines and vehicles;
step 2.2: traversing the point clouds and calculating the average distance between each point cloud and n neighborhood points of the point cloud; then, calculating standard deviation of average distance between the point cloud and the neighborhood pointσAnd mean valueμDistance thresholdd max The calculation formula is thatd max =μ+α×σWhereinαIs a scaling factor, i.e., the number of domain points; finally, traversing the point cloud again and eliminating the point cloud with the average distance larger thand max Is a point of (2);
step 3: extracting structural surface point clouds from point cloud data containing normal vectors by adopting a RANSAC method, establishing a structural surface point cloud set, optimizing parameters of the structural surface point clouds by adopting a nonlinear model method for the point clouds in the structural surface point cloud set, and obtaining an optimized structural surface point cloud data set;
step 4: and identifying the structural surface according to the optimized structural surface point set by using a region growing algorithm, and calculating the tendency and the inclination angle of the structural surface.
2. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in claim 1, wherein the step of establishing the point cloud data index by using kd-tree method in step 2 comprises:
step 2.3: sorting the preprocessed point cloud on three coordinate axes, respectively calculating standard deviation of preprocessed point cloud data on three dimensions, sorting the preprocessed point cloud data according to the standard deviation, and taking the dimension with the largest standard deviation as a first sorting coordinate axis, wherein the sequence of the standard deviation is the replacement sequence of the sorting coordinate axis;
step 2.4: calculating the median of all the point cloud data on three sequencing coordinate axes, putting the median on a current dividing node to serve as a root node of a binary tree, dividing point clouds smaller than or equal to the median into a left subtree of the binary tree, and dividing point clouds larger than the median into a right subtree of the binary tree;
step 2.5: after the sub-division is completed, the sorting coordinate axes are replaced according to the sequence in the step 2.3;
step 2.6: repeating the steps 2.4 to 2.5 until the establishment of the kd-tree index of the point cloud data is completed.
3. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in claim 1, wherein the step 3 is specifically as follows:
step 3.1: randomly extracting 4 non-collinear sample data from the point cloud data containing normal vectors, calculating a transformation matrix H, marking the transformation matrix H as a model M, and calculating a matrix H calculation formula as followsWherein (x, y) represents the position of the corner of the target image, (x ', y') is the position of the corner of the scene image, and s is the scale parameter;
step 3.2: calculating projection errors of all point cloud data containing normal vectors and a model M, and if the projection errors are smaller than a threshold value, adding the projection errors into an inner point set Q, wherein the threshold value is determined according to the size of data noise;
step 3.3: when the number of elements in the inner point set Q is greater than the optimal inner point set Q-best, then update Q-best=q, and update the maximum number of iterations at the same timeK,WhereinpIn order for the confidence level to be high,wthe proportion of the elements contained in the inner point set Q to the elements in all the point cloud data containing normal vectors;
step 3.4: if the iteration number is greater than K, exiting; otherwise the iteration number is increased by 1 and steps 3.1 to 3.3 are repeated.
4. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in claim 3, wherein: and 3.1, taking the scale parameter as the standard deviation of the point cloud data.
5. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in claim 3, wherein: the threshold value in step 3.2 is taken to be 0.01.
6. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in claim 3, wherein: and 3.3, taking 0.995 for confidence level p in the step.
7. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in any one of claims 1 to 6, wherein the step 4 is specifically as follows:
step 4.1: randomly selecting one point in the point cloud area as an initial point;
step 4.2: calculating the included angle of the normal vector of the initial point and the included angles of the normal vectors of all points in the neighborhood around the initial point;
step 4.3: if the absolute value of the difference between the points in the vicinity of the initial point and the normal vector angle of the initial point is smaller than the flatness constraint thresholdεMerging the point into a point cloud area where the initial point is located;
step 4.4: searching a new initial point in the points merged into the point cloud area until no new point is merged into the point cloud area, so as to obtain a merging set of all the point cloud areas;
step 4.5: and fitting all points in each merging set to obtain the normal line of the plane of the structural surface, identifying the structural surface and calculating the inclination and the inclination angle of the structural surface.
8. The rapid identification method of laser point cloud rock mass structure based on RANSAC algorithm as claimed in claim 7, wherein: in step 4.3εThe value is 0.1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410153806.8A CN117689965A (en) | 2024-02-04 | 2024-02-04 | Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410153806.8A CN117689965A (en) | 2024-02-04 | 2024-02-04 | Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117689965A true CN117689965A (en) | 2024-03-12 |
Family
ID=90128658
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410153806.8A Pending CN117689965A (en) | 2024-02-04 | 2024-02-04 | Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117689965A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180356213A1 (en) * | 2016-09-14 | 2018-12-13 | Hangzhou Scantech Co., Ltd | Three-dimensional sensor system and three-dimensional data acquisition method |
CN109658431A (en) * | 2018-12-26 | 2019-04-19 | 中国科学院大学 | Rock mass point cloud plane extracting method based on region growing |
CN109685821A (en) * | 2018-12-26 | 2019-04-26 | 中国科学院大学 | Region growing 3D rock mass point cloud plane extracting method based on high quality voxel |
CN115439839A (en) * | 2022-08-04 | 2022-12-06 | 三峡大学 | Multi-type structural surface layered identification method based on point cloud data |
-
2024
- 2024-02-04 CN CN202410153806.8A patent/CN117689965A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180356213A1 (en) * | 2016-09-14 | 2018-12-13 | Hangzhou Scantech Co., Ltd | Three-dimensional sensor system and three-dimensional data acquisition method |
CN109658431A (en) * | 2018-12-26 | 2019-04-19 | 中国科学院大学 | Rock mass point cloud plane extracting method based on region growing |
CN109685821A (en) * | 2018-12-26 | 2019-04-26 | 中国科学院大学 | Region growing 3D rock mass point cloud plane extracting method based on high quality voxel |
CN115439839A (en) * | 2022-08-04 | 2022-12-06 | 三峡大学 | Multi-type structural surface layered identification method based on point cloud data |
Non-Patent Citations (2)
Title |
---|
SANGMIN OH 等: "Building Component Detection on Unstructured 3D Indoor Point Using RANSAC-Based Region Growing", 《MDPI》, 30 June 2021 (2021-06-30) * |
高梓成: ""基于TLS的三峡库区典型试点崩滑灾害监测调查研究"", 《中国优秀硕士学位论文全文数据库 基础科学辑 (月刊)》, 15 July 2020 (2020-07-15), pages 3 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106780524B (en) | Automatic extraction method for three-dimensional point cloud road boundary | |
Morgan et al. | Automatic building extraction from airborne laser scanning data | |
Briese | Three-dimensional modelling of breaklines from airborne laser scanner data | |
CN111553292B (en) | Rock mass structural plane identification and occurrence classification method based on point cloud data | |
Boulaassal et al. | Automatic extraction of planar clusters and their contours on building façades recorded by terrestrial laser scanner | |
CN105136054A (en) | Fine structure deformation monitoring method and system based on ground three-dimensional laser scanning | |
CN114332348B (en) | Track three-dimensional reconstruction method integrating laser radar and image data | |
CN114419085A (en) | Automatic building contour line extraction method and device, terminal device and storage medium | |
CN115564926B (en) | Three-dimensional patch model construction method based on image building structure learning | |
CN111340723A (en) | Terrain-adaptive airborne LiDAR point cloud regularization thin plate spline interpolation filtering method | |
Ji et al. | A coarse-to-fine strip mosaicing model for airborne bathymetric LiDAR data | |
Sun et al. | An automatic coordinate unification method of multitemporal point clouds based on virtual reference datum detection | |
Zhu et al. | Triangulation of well-defined points as a constraint for reliable image matching | |
CN113409332A (en) | Building plane segmentation method based on three-dimensional point cloud | |
CN117689965A (en) | Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm | |
CN115965712A (en) | Building two-dimensional vector diagram construction method, system, equipment and storage medium | |
CN115131571A (en) | Building local feature point identification method based on six fields of point cloud pretreatment | |
CN113567968B (en) | Underwater target real-time segmentation method based on shallow water multi-beam water depth data and application thereof | |
He et al. | Rapid assessment of slope deformation in 3D point cloud considering feature-based simplification and deformed area extraction | |
Sester et al. | Reconstruction of building ground plans from laser scanner data | |
Sun et al. | Datum feature extraction and deformation analysis method based on normal vector of point cloud | |
Toth et al. | Strip adjustment | |
Yang et al. | A least square matching optimization method of low altitude remote sensing images based on self-adaptive patch | |
Lee et al. | Effective Reduction of Horizontal Error in Laser Scanning Information by Strip‐Wise Least Squares Adjustments | |
CN117854060B (en) | Tunnel rock body planar crack identification method and system based on deep learning |
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 |