CN116469092A - Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm - Google Patents
Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm Download PDFInfo
- Publication number
- CN116469092A CN116469092A CN202211504717.0A CN202211504717A CN116469092A CN 116469092 A CN116469092 A CN 116469092A CN 202211504717 A CN202211504717 A CN 202211504717A CN 116469092 A CN116469092 A CN 116469092A
- Authority
- CN
- China
- Prior art keywords
- points
- contour
- point
- algorithm
- extraction
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 41
- 230000011218 segmentation Effects 0.000 claims abstract description 18
- 238000001914 filtration Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 abstract description 7
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000010276 construction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a self-adaptive Alpha Shapes contour extraction method based on a DBSCAN algorithm, which can realize contour high-quality and high-precision extraction. According to the invention, the ROR filtering algorithm and the minimum point constraint VoxelGrid voxel filtering algorithm are combined to remove the miscellaneous points in the initial contour points, so that the contour lines are further regular, noise point rejection is realized, the initial contour points are obtained through denoising, the boundary is subjected to rapid clustering segmentation based on the DBSCAN algorithm accelerated by the KD tree, and the self-adaptive AS algorithm is established to combine the least square linear fitting precision advantage and AS algorithm traversal circle detection contour point optimization. The problem that the contour lines in the AS algorithm are deformed or zigzag is effectively and rapidly removed, the contour line extraction precision of the three-dimensional building is improved, and the contour of the building with more corner features can be accurately and rapidly extracted from the laser point cloud.
Description
Technical Field
The invention relates to the technical field of urban three-dimensional modeling of laser radar sensors, in particular to a self-adaptive Alpha Shapes contour extraction method based on a DBSCAN algorithm.
Background
Along with the great reduction of the cost of the laser radar, it is becoming very common to acquire high-resolution laser point cloud information by utilizing the laser radar technology, and the laser point cloud information has become an accurate data support for urban three-dimensional model establishment, high-precision map construction and intelligent management of a digital park, and has been widely applied to the fields of mobile mapping, cultural relic reconstruction, unmanned driving, forestry estimation, disaster monitoring and the like. As the laser radar technology is not well known, the research of the related point cloud data processing theory is still in a preliminary stage, and although certain achievements are achieved in the aspects of point cloud preprocessing, registration, segmentation, modeling and the like, many problems in the aspect of building contour target feature extraction are not completely solved, and the high-precision extraction algorithm theory of building contour line features needs to be further researched. The contour information of the building comprises boundaries between wall surfaces, contours of a roof, boundaries of a balcony and the like, the contour information is very critical to the salient of important features on the surface of the building, the existing contour line feature extraction algorithm mainly comprises two major types of image-based extraction algorithms and point cloud-based extraction algorithms, wherein the image-based extraction algorithms mainly adopt overlapping images to meet to generate point clouds so as to extract the contour of the building, the point cloud-based extraction algorithms directly process the point clouds, and the discrete point cloud sets are used for extracting the visual shape of a target, and although the image-based extraction algorithms can ensure the extraction accuracy, the algorithm extraction procedures are complex, the period is long, and the result is influenced by the construction accuracy of a three-dimensional model; however, although the extraction algorithm based on the point cloud has high automation degree and high extraction efficiency, the currently widely applied Alpha Shapes extraction algorithm is easy to be interfered by noise points to cause saw-tooth deformation of the profile, because the neighborhood search radius of European cluster segmentation is fixed, the phenomenon of under segmentation occurs along with the larger or smaller radius, and the noise points and the characteristic points are segmented according to the convergence principle, so that the segmentation results of other points are seriously affected by the noise points.
Disclosure of Invention
In view of the above, the invention provides a self-adaptive Alpha Shapes contour extraction method based on a DBSCAN algorithm, which can realize high-quality and high-precision contour extraction.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
a self-adaptive Alpha Shapes contour extraction method based on a DBSCAN algorithm comprises the following steps:
step 1, acquiring an initial contour point, and then removing miscellaneous points in the initial contour point by adopting a combination of a ROR filtering algorithm and a VoxelGrid voxel filtering algorithm with minimum point constraint;
step 2, clustering and segmenting the initial contour points by adopting a DBSCAN segmentation algorithm based on KD trees to obtain segmented contour point subsets;
step 3, sequentially aiming at each segmented profile point subset, acquiring profile points by using a self-adaptive Alpha Shapes method, wherein the specific process is as follows: randomly selecting some points from the extracted contour points, fitting the contour points by using the points by adopting a least square model to obtain a linear model, calculating a judgment radius R based on the positions of neighbor points of the linear model, drawing a circle with the radius R by using any two points, and if no other data points exist in any one circle, considering the boundary points of the two points, wherein the connecting line is a boundary line segment; traversing all the segmented profile point subsets to obtain a final boundary, and finishing profile extraction.
In the step 2, a DBSCAN segmentation algorithm based on a KD tree is adopted to perform clustering segmentation on the initial contour points, and a segmented contour point subset is obtained, and the specific steps are as follows:
reading the denoised contour point cloud, and setting a threshold epsilon 1 Optionally, a point m is selected, the KD tree is utilized to quickly find the field epsilon of the point, if the distance point m is not divided into any core points of any class, then all epsilon of the point m is accessed by finding all points which are reachable from the density of m to form a region containing m 1 Is a point of (2); if the points are not already presentAssigned a cluster, then the new cluster tags just created are assigned to them; if they are core samples, then access is made sequentially, and so on; the clusters gradually increase until epsilon in the cluster 1 No more core samples within the distance; all points are accessed to obtain a segmented profile point subset.
In step 1, three different densities are set for the initial contour points in the Cloud computer, three kinds of data are extracted, and the extraction precision and the time used for each algorithm are recorded to ensure the optimal configuration.
In the step 1, three-dimensional R-tree in PCL library is adopted to realize roof initial contour point extraction.
Wherein in the step 3, the radius d is set min o(o x ,o y ) Substituting the circle center coordinates of the circles of the two points to calculate the judgment radius R based on the positions of the neighbor points of the linear model, and passing through any two points P 1 (x 1 ,y 1 )、P 2 (x 2 ,y 2 ) Drawing a circle with radius R, wherein the calculation formula is as follows:
wherein,,wherein d is min Is the minimum distance from all contour points to a straight line in the field of straight line equations.
Advantageous effects
1. According to the self-adaptive Alpha Shapes contour extraction method based on the DBSCAN algorithm, the PCL (Point Cloud Library) library is called to preprocess the initial point cloud, so that projection coordinate calculation of the point cloud is realized; removing the miscellaneous points in the initial contour points by combining a ROR filtering algorithm and a minimum point constraint VoxelGrid voxel filtering algorithm, so that the contour lines are further regular, noise point elimination is realized, the initial contour points are obtained through denoising, the boundary is subjected to rapid clustering segmentation based on a DBSCAN algorithm accelerated by KD trees, and the linear fitting precision advantage of a self-adaptive AS algorithm combined with least square and AS algorithm traversal circle detection contour point optimization are established. The method is insensitive to noise, can effectively judge outliers, and automatically divide the outliers into a cluster, so that the noise points have no influence on segmentation, a better segmentation result can be obtained, the problems that the classical Alpha Shapes contour automatic extraction algorithm is easily influenced by the noise points to cause the contour to deform or be in a saw-tooth shape, the extraction precision is low and the like are solved, the problem that contour lines in an AS algorithm deform or are in a saw-tooth shape is effectively and rapidly removed, the contour line extraction precision of a three-dimensional building is improved, and the contour of a building with more corner features can be accurately and rapidly extracted from a laser point cloud.
2. In the invention, three different densities are set for the initial contour points in the Cloud computer, three kinds of data are extracted by the method, and the extraction precision and the time used by each algorithm are recorded to ensure the optimal configuration.
3. In the invention, the adjacent topological relation between the three-dimensional R-tree point cloud construction data is adopted, so that the point cloud boundary can be quickly and accurately obtained, the internal topological relation between the surfaces is considered, the multi-top-surface outline of the multi-level building can be effectively extracted, and the outline extraction quality precision is improved.
Drawings
Fig. 1 is a schematic diagram of a process for obtaining a building initial contour point according to the present invention.
Fig. 2 is a schematic diagram of a contour point optimization flow according to an embodiment of the present invention.
Detailed Description
The invention will now be described in detail by way of example with reference to the accompanying drawings.
Because the neighborhood search radius of European cluster segmentation is fixed, the phenomenon of under segmentation occurs along with larger or smaller radius, and noise points and characteristic points are segmented according to the convergence principle, so that the noise points can seriously influence the segmentation results of other points. The Alpha Shapes algorithm has redundant corner points in the extraction process because the extracted roof points contain other points.
Aiming at the phenomenon that the existing building contour extraction algorithm is easily influenced by noise points to cause contour deformation or the phenomenon that the contour is saw-tooth due to the fact that the contour points are broken or missing, the invention provides an adaptive Alpha Shapes contour extraction method based on an acceleration DBSCAN. Firstly, adopting a three-dimensional R-tree to construct a proximity topological relation between data, rapidly and accurately obtaining a point cloud boundary, then carrying out rapid clustering segmentation on the boundary based on a DBSCAN algorithm accelerated by a KD tree, carrying out contour point optimization after obtaining contours of different clusters, and particularly establishing self-adaptive Alpha Shapes algorithm and traversing circle detection contour point optimization by combining the least square linear fitting precision advantage and the Alpha Shapes algorithm. The problem that the contour lines deform or are in a saw-tooth shape in the Alpha Shapes algorithm is effectively and rapidly removed, saw-tooth Shapes of the contour lines can be removed, and the contour lines at the grooves and the protrusions can be well identified. The method specifically comprises the following steps:
step 1, an initial contour point is obtained, and the flow is shown in fig. 1, and specifically comprises the following steps:
acquiring a building roof point cloud and completing projection coordinate calculation of the point cloud, and then calling a PCL (Point Cloud Library) library to preprocess the initial point cloud, specifically, adopting a three-dimensional R-tree in a PCL library to realize roof initial contour point extraction: the boundary points of the two-dimensional point cloud are usually located at the outermost periphery of the point cloud, and the points near the boundary points are mostly located on the same side of the boundary points.
Further, three different densities are set for the initial contour points in the Cloud computer, the three data are extracted by the method, the extraction precision and the time used for each algorithm are recorded, and the optimal configuration is ensured.
Although the three-dimensional R-tree algorithm is used to extract the initial boundaries of the building, there are many outliers in the extracted contour points that will affect the contour of the building. And then removing the miscellaneous points in the initial contour points by combining the ROR filtering algorithm and the VoxelGrid voxel filtering algorithm with the minimum point constraint so as to further normalize the contour lines and remove noise points.
The contour point optimization flow is shown in fig. 2, and is specifically as follows
And 2, clustering and segmenting the initial contour points by adopting a DBSCAN segmentation algorithm based on a KD tree to obtain a segmented contour point subset, and providing key support for follow-up high-efficiency and automatic contour point optimization. The method comprises the following specific steps:
reading the denoised contour point cloud, and setting a threshold epsilon 1 Optionally, a point m is selected, the KD tree is utilized to quickly find the field epsilon of the point, if the distance point m is not divided into any core points of any class, then all epsilon of the point m is accessed by finding all points which are reachable from the density of m to form a region containing m 1 Is a point of (2); if the points have not been assigned a cluster, then a new cluster label that has just been created is assigned to them; if they are core samples, then access is made sequentially, and so on; the clusters gradually increase until epsilon in the cluster 1 No more core samples within the distance; all points are accessed to obtain a segmented profile point subset P {1}, P {2}, … P { n }.
Step 3, sequentially aiming at each segmented profile point subset, acquiring profile points by using a self-adaptive Alpha Shapes method, wherein the specific process is as follows: randomly selecting some points from the extracted contour points, fitting the contour points by using a least square model to obtain a linear model, calculating a discrimination radius R based on the positions of neighbor points of the linear model, and passing through any two points P 1 、P 2 Drawing circles with radius R, and if no other data points exist in any one circle, considering the point P 1 、P 2 Is a boundary point, its connection line P 1 P 2 And traversing all the segmented profile point subsets to obtain a final boundary for the boundary line segment, and finishing profile extraction. The specific implementation process is as follows:
setting a threshold epsilon 2 Count value theta i =0, randomly selecting two points in a subset P { n }, calculating a linear equation, and obtaining the distances d from all the points of the subset to the line, if d>ε 2 Theta is then i ＝θ i +1, otherwise, selecting two points in a subset P { n }, calculating a linear equation until d>ε 2 ；
The most recorded straight line set is a fitting straight line, and the distance d between all contour line points and the straight line in the field epsilon of the straight line equation is calculated 2 And record the minimum value as d min ；
The calculation has any twoThe center of the circle of points will have radius d min o(o x ,o y ) Substituting to calculate the center coordinates of the circles of the two points, if the distance from the center d min If no other points exist in the length range, judging the two points as contour points and adding the contour points into the point set S;
and repeatedly calculating n X (n-1) sides, connecting contour points, and generating a contour line.
Further, fitting contour points by adopting a least square model, and obtaining a straight line model by the following specific modes:
the point-wise equation for a spatial line given a known point in space is:
wherein l, n, p are direction vectors of a space straight line, x 0 、y 0 、z 0 Is a known point of the spatial straight line.
Calculating any point P (x, y, z) to a spatial point Q in a space 1 (x 1 ,y 1 ,z 1 ) And Q 2 (x 2 ,y 2 ,z 2 ) The space straight line distance of the components isThe point-wise equation is deformed into: />Let-> The point-wise equation is expressed as: />
The three-dimensional space straight line fitting method adopting the least square criterion is used for spatially calculating four geometric parameters converted from six geometric parameters of a straight line, and calculating an approximate value of x:
calculating the approximation value of x and the actual measurement value of x i Squaring the sum of the differences:
approximating y:
calculating the approximation value and the actual measurement value y of y i Squaring the sum of the differences:
the derivative was taken with respect to a, b, c, d:
obtaining Q based on least squares criterion x And Q y Is the minimum value of (2):
calculating and judging radius R based on positions of neighborhood points of linear model and passing through any two points P 1 (x 1 ,y 1 )、P 2 (x 2 ,y 2 ) Drawing a circle with radius R, wherein the calculation formula is as follows:
wherein,,if there are no other data points in any one circle, then point P is considered 1 、P 2 Is a boundary point, its connection line P 1 P 2 Is a boundary line segment.
In summary, the above embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (5)
1. The self-adaptive Alpha Shapes contour extraction method based on the DBSCAN algorithm is characterized by comprising the following steps of:
step 1, acquiring an initial contour point, and then removing miscellaneous points in the initial contour point by adopting a combination of a ROR filtering algorithm and a VoxelGrid voxel filtering algorithm with minimum point constraint;
step 2, clustering and segmenting the initial contour points by adopting a DBSCAN segmentation algorithm based on KD trees to obtain segmented contour point subsets;
step 3, sequentially aiming at each segmented profile point subset, acquiring profile points by using a self-adaptive Alpha Shapes method, wherein the specific process is as follows: randomly selecting some points from the extracted contour points, fitting the contour points by using the points by adopting a least square model to obtain a linear model, calculating a judgment radius R based on the positions of neighbor points of the linear model, drawing a circle with the radius R by using any two points, and if no other data points exist in any one circle, considering the boundary points of the two points, wherein the connecting line is a boundary line segment; traversing all the segmented profile point subsets to obtain a final boundary, and finishing profile extraction.
2. The method of claim 1, wherein in the step 2, the initial contour points are clustered and segmented by using a DBSCAN segmentation algorithm based on KD-tree to obtain the segmented contour point subsets, and the specific steps are as follows:
reading the denoised contour point cloud, and setting a threshold epsilon 1 Optionally, a point m is selected, the KD tree is utilized to quickly find the field epsilon of the point, if the distance point m is not divided into any core points of any class, then all epsilon of the point m is accessed by finding all points which are reachable from the density of m to form a region containing m 1 Is a point of (2); if the points have not been assigned a cluster, then a new cluster label that has just been created is assigned to them; if they are core samples, then access is made sequentially, and so on; the clusters gradually increase until epsilon in the cluster 1 No more nuclei within the distanceUntil a heart sample; all points are accessed to obtain a segmented profile point subset.
3. The method according to claim 1 or 2, wherein in step 1, three different densities are set for the initial contour points in the Cloud computer, the three data are extracted, and the extraction accuracy and the time used for each algorithm are recorded to ensure the optimal configuration.
4. The method according to claim 1 or 2, wherein in the step 1, roof initial contour point extraction is implemented by using a three-dimensional R-tree in a PCL library.
5. The method according to claim 1 or 2, wherein in step 3, the radius d is set to min o(o x ,o y ) Substituting the circle center coordinates of the circles of the two points to calculate the judgment radius R based on the positions of the neighbor points of the linear model, and passing through any two points P 1 (x 1 ,y 1 )、P 2 (x 2 ,y 2 ) Drawing a circle with radius R, wherein the calculation formula is as follows:
wherein,,wherein d is min Is the minimum distance from all contour points to a straight line in the field of straight line equations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211504717.0A CN116469092A (en) | 2022-11-28 | 2022-11-28 | Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211504717.0A CN116469092A (en) | 2022-11-28 | 2022-11-28 | Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116469092A true CN116469092A (en) | 2023-07-21 |
Family
ID=87177594
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211504717.0A Pending CN116469092A (en) | 2022-11-28 | 2022-11-28 | Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116469092A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117934324A (en) * | 2024-03-25 | 2024-04-26 | 广东电网有限责任公司中山供电局 | Denoising method and device for laser point cloud data and radar scanning device |
-
2022
- 2022-11-28 CN CN202211504717.0A patent/CN116469092A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117934324A (en) * | 2024-03-25 | 2024-04-26 | 广东电网有限责任公司中山供电局 | Denoising method and device for laser point cloud data and radar scanning device |
| CN117934324B (en) * | 2024-03-25 | 2024-06-11 | 广东电网有限责任公司中山供电局 | Denoising method and device for laser point cloud data and radar scanning device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109961440B (en) | Three-dimensional laser radar point cloud target segmentation method based on depth map | |
| CN112070769B (en) | Layered point cloud segmentation method based on DBSCAN | |
| WO2024077812A1 (en) | Single building three-dimensional reconstruction method based on point cloud semantic segmentation and structure fitting | |
| CN110222642B (en) | Plane building component point cloud contour extraction method based on global graph clustering | |
| Lari et al. | An adaptive approach for the segmentation and extraction of planar and linear/cylindrical features from laser scanning data | |
| CN113920134B (en) | Slope ground point cloud segmentation method and system based on multi-line laser radar | |
| CN112396641B (en) | Point cloud global registration method based on congruent two-baseline matching | |
| Cheng et al. | Building boundary extraction from high resolution imagery and lidar data | |
| CN113012161B (en) | Stacked scattered target point cloud segmentation method based on convex region growth | |
| Wei et al. | Automatic coarse registration of point clouds using plane contour shape descriptor and topological graph voting | |
| CN114926699A (en) | Indoor three-dimensional point cloud semantic classification method, device, medium and terminal | |
| CN113640826B (en) | Obstacle identification method and system based on 3D laser point cloud | |
| CN111783722B (en) | Lane line extraction method of laser point cloud and electronic equipment | |
| CN111968224A (en) | Ship 3D scanning point cloud data processing method | |
| CN114821571A (en) | Point cloud processing method for power cable identification and reconstruction | |
| CN116109601A (en) | Real-time target detection method based on three-dimensional laser radar point cloud | |
| CN116469092A (en) | Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm | |
| CN115861247A (en) | High-resolution remote sensing image contour multistage regularization method, system and application | |
| CN116258857A (en) | Outdoor tree-oriented laser point cloud segmentation and extraction method | |
| CN113409332B (en) | Building plane segmentation method based on three-dimensional point cloud | |
| WO2024125434A1 (en) | Regional-consistency-based building principal angle correction method | |
| CN114325760A (en) | Road tunnel quality inspection obstacle avoidance method and system based on multi-line laser radar | |
| CN117253205A (en) | Road surface point cloud rapid extraction method based on mobile measurement system | |
| CN114820505A (en) | Non-contact measurement method of dynamic target | |
| CN114283199B (en) | Dynamic scene-oriented dotted line fusion semantic SLAM method |
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 |