CN109685821A - Region growing 3D rock mass point cloud plane extracting method based on high quality voxel - Google Patents
Region growing 3D rock mass point cloud plane extracting method based on high quality voxel Download PDFInfo
- Publication number
- CN109685821A CN109685821A CN201811603927.9A CN201811603927A CN109685821A CN 109685821 A CN109685821 A CN 109685821A CN 201811603927 A CN201811603927 A CN 201811603927A CN 109685821 A CN109685821 A CN 109685821A
- Authority
- CN
- China
- Prior art keywords
- plane
- voxel
- point cloud
- point
- growing
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Remote Sensing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Graphics (AREA)
- Image Generation (AREA)
Abstract
The region growing 3D rock mass point cloud plane extracting method based on high quality voxel that the present invention relates to a kind of.The present invention mainly includes 3 stages；(1) voxel-based space is divided detects with coplanarity；(2) the rock mass point cloud initial plane based on algorithm of region growing is extracted；(3) the planar growth sheet based on neighborhood relationships.Firstly, initial rock mass point cloud is divided into voxel grid, the reliable region growing unit of coplanarity Detection and Extraction is carried out in each voxel inner utilization stochastical sampling consistency, while establishing neighborhood index；Then region growing is carried out using the growing element of extraction and neighborhood index obtain initial plane set；Diauxic growth is carried out to the left point cloud in each plane surrounding neighbors using neighborhood relationships finally for initial plane set, extracts final plane set.The characteristic plane that the present invention is suitable for large scene rock laser scanning point cloud extracts, and can efficiently, accurately extract the plane in rock mass point cloud.
Description
Technical field
The invention belongs to the crossing domains that computer vision and three dimensional point cloud are handled, and are related to a cloud feature extraction, rock
The technologies such as body three-dimensional reconstruction, in particular to the region growing 3D rock mass point cloud plane based on high quality voxel are extracted.
Background technique
Plane extraction in point cloud is the important content in many computer graphics, image procossing and computer vision,
Including three-dimensional reconstruction, object identification, virtual reality etc..Currently, laser scanner can be with the acquisition of high-resolution, high accuracy
The coordinate information of the three-dimensional space of target is further promoted for three-dimensional point cloud processing and the research of modeling process, and point
Plane extracts initial step as points cloud processing and analysis in cloud, have become points cloud processing field research hotspot it
One.
Current many researchers propose a large amount of plane extracting method, wherein sampling consistency (Random immediately
Sample Consensus, RANSAC), Hough transformation (Hough Transform, HT) and region growing (Region
Growing, RG) it is most normal adopted three kinds of typical methods.But traditional stochastical sampling consistency algorithm, RANSAC have
Preferable robustness, the degree of automation is higher, and can also show preferable performance in the scene of large scale, but it is essential
On be a greedy algorithm, can include by adjacent boundary point or intersection point from being extracted in cloud in maximum planarization process every time
Into plane；Hough transformation is capable of the detection plane of robust, but its computation complexity it is excessively high be the algorithm main problem.Region
Growth scans for part, identifies and expands the region with same characteristic features, but its growth course is by seed point and life
Long rule is affected.
In rock mass engineering project field, rock mass surface is often made of irregular and different degree of roughness plane, to these
The analysis and calculating of plane play an important role to the stability of analysis rock mass.But the plane extraction in point cloud is mostly at present
Research to urban architecture object point cloud, this kind of 3D laser point cloud include mostly rule geometrical characteristic face (such as building wall wall surface,
Ground and desktop etc.), plane is more smooth, and rock mass point cloud uneven surface, coarse, and it is in irregular shape, it is mostly with high gradient slope
Form present, differ greatly with common city point cloud.Therefore the characteristic face of rock mass point cloud is extracted and three-dimensional modeling becomes rock
The pith of body engineering.
Summary of the invention
The present invention shows the characteristics such as coarse, in irregular shape for rock mass surface, in order to overcome conventional planar to extract skill
Art rock mass point cloud plane extraction in low efficiency, accuracy rate is low, there are over-segmentation and less divideds the problems such as, propose one kind and be based on
The region growing 3D rock mass point cloud plane extracting method of high quality voxel can be done and realize quick plane extraction algorithm, and have
Higher accuracy.
The present invention provides a kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel, mainly forgives
Following steps:
Step 1, voxel-based space are divided to be detected with coplanarity: the normal vector of each point in first choice estimation point cloud；So
Voxel is quickly established according to the bounding box size of rock mass point cloud afterwards；It is total in each voxel inner utilization stochastical sampling consistency
Face property is detected to establish growing element (plane)；Neighborhood index finally is established to each growing element.
Step 2, the rock mass point cloud initial plane based on algorithm of region growing are extracted: according to the growing element after foundation, benefit
Initial plane set is extracted with voxel-based algorithm of region growing；
Step 3, the planar growth based on neighborhood relationships: it on the basis of each plane after extracting, establishes point cloud and merges standard
Then, the left point cloud after extraction is determined in the neighborhood of its plane, obtains final plane set.
A kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel of the present invention, as excellent
Mode is selected, step 1 further includes steps of
Step 1.1 utilizes the normal vector of each point in principal component analysis (PCA) estimation point cloud.For any one in cloud
Point p, radius be r region in total include k point.Then this k point constructs covariance matrix Σ, wherein λ1< λ2< λ3
Indicate three characteristic values of covariance matrix Σ, v1,v2,v3For its corresponding feature vector.Then its minimal eigenvalue λ1It is corresponding
Feature vector v1For the normal vector of p point.
Σ·vj=λj·vj,j∈{1,2,3}
The given voxel side length of step 1.2, basis, will put cloud rapid voxel.
Step 1.3, for each voxel inside, if count be greater than certain threshold value (N_r) if using be based on normal vector
Stochastical sampling consistency (ND-RANSAC) carry out coplanarity detection, for extracting growing element (plane).It is assumed that each voxel
N number of point is contained in inside, therefrom randomly selects 3 non-coplanar points and establishes plane pl, is carried out according to remaining cloud to this plane
Marking, if the threshold value d that the distance of point to plane is less than1And the normal direction of point and the angle of plane normal direction are less than threshold θ1Then score adds
1；This process iteration T times, the maximum plane of score are final growing element (plane)；
Step 1.4, according to the growing element of foundation on the basis of, to each growing element establish neighborhood index.
A kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel of the present invention, as excellent
Mode is selected, step 2 further includes steps of
Step 2.1 is therefrom chosen the growing element that step 1 is extracted and is put down comprising the most plane of point number as seed
Face；
Step 2.2 establishes region growing criterion: the normal angle of 1 liang of plane is less than certain threshold θ2；2 liang of planes
Distance is less than d2。
Step 2.3, centered on seed plane, indexed using the established neighborhood of step 1.4, its peripheral neighborhood carried out
Search；The growing element for meeting growth criterion is merged；The unit being newly added continues to give birth to as new seed plane
It is long, until not new plane addition, then a complete plane is extracted from cloud.
Step 2.4, the process iteration of step 2.1-2.3 for several times, until all growing elements are all processed, obtain
To initial plane set.
A kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel of the present invention, as excellent
Mode is selected, step 3 further includes steps of
Step 3.1, the initial plane set extracted every time from step 2 choose the maximum plane P that counts；
Step 3.2, from the left point cloud after extraction, select P neighborhood and inside left point converge close R；
Step 3.3 establishes point cloud merging criterion: if 1 point is less than certain threshold θ apart from normal angle to plane P3；2
Point is less than d to plane P distance3.Point each in R is determined, the point for meeting merging criterion is added in plane P, is laid equal stress on
New estimation plane parameter；
Step 3.4, repeat the above steps for each initial plane that step 2 is extracted 3.1-3.4, obtains final plane
Set.
The present invention is directed to the analysis of rock mass point cloud characteristic, substantially increases algorithm using the region growing based on extraction and imitates
Rate carries out coplanarity detection using stochastical sampling consistency, provides reliable growing element for area growth process, improve calculation
The accuracy of method.Left point cloud is handled using neighborhood relationships simultaneously, algorithm is improved and is coping with irregular and degree of roughness not
One rock mass environment, also improves the accuracy of algorithm.
Detailed description of the invention
Fig. 1 is the region growing 3D rock mass point cloud plane extracting method flow chart based on high quality voxel.
The growing element schematic diagram established after the detection of Fig. 2 coplanarity.
Fig. 3 is that index establishes regular schematic diagram.
Fig. 4 is voxel-based region growing result.
Fig. 5 is that the plane based on neighborhood information extracts result
Specific embodiment
Present invention is primarily based on computer vision and points cloud processing technology, propose that a kind of region based on high quality voxel is raw
Long 3D rock mass point cloud plane extracting method.The present invention comprehensively utilizes stochastical sampling consistency and voxel-based region growing carries out
Characteristic face extracts, and improves the efficiency and accuracy of plane extraction, extends the scope of application of existing plane extracting method, can
Accurate plane is obtained under the complex scenes such as rock mass extracts result.
In order to keep the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with drawings and examples, to this hair
It is bright to be described in further detail.Method provided by the invention can use computer software technology implementation process, overall technology process
Figure is referring to Fig. 1, comprising the following steps:
Step 1, in order to realize that the detection of reliable coplanarity and neighborhood divide, it is necessary first to set normal estimation radius and
The specific implementation process of the side length of voxel, example is described as follows:
The given normal estimation radius r (value 0.3m in this example) of step 1.1, basis, carries out normal estimation.Benefit
With the normal vector of each point in principal component analysis (PCA) estimation point cloud.For any point p in cloud, radius is the region of r
In in total include k point.Then this k point constructs covariance matrix Σ, wherein λ1< λ2< λ3Indicate the three of covariance matrix Σ
A characteristic value, v1,v2,v3For its corresponding feature vector.Then its minimal eigenvalue λ1Corresponding feature vector v1For the normal direction of p point
Amount；
Σ·vj=λj·vj,j∈{1,2,3}
The given voxel side length voxel_size (value 0.8 in this example) of step 1.2, basis, will put cloud voxelization；
Step 1.3 for each voxel inside, if points be greater than certain threshold value Nr(value 30 in this example) is then sharp
Coplanarity detection is carried out with the stochastical sampling consistency (ND-RANSAC) based on normal vector, for extracting growing element (plane)
It is assumed that containing N number of point inside each voxel, therefrom randomly selects 3 non-coplanar points and establish plane pl, according to remaining cloud
It gives a mark to this plane, if the threshold value d that the distance of point to plane is less than1The normal direction and put down that (value 0.08m in this example) is put
The angle of face normal direction is less than threshold θ1(10 ° of value in this example) then score adds 1；This process iteration T times, the maximum plane of score
For final growing element (plane)；
Step 1.4 determines the number of iterations T, it is assumed that contains N number of point, maximum planes p wherein included in each voxelmaxContain
There is n point, needs to carry out T iteration to extract this maximum planes.Go out p by one-time detectionmaxProbability be p1, lead to
Crossing the successful probability of T iterative detection is p2, ptFor by T iterative detection a to pmaxMinimum probability；
Step 1.5, the result after RANSAC carries out coplanarity detection are as shown in Figure 2.Each growing element is built
Vertical neighborhood index.Shown in Fig. 3, in three-dimensional space for each growing element (central area), around be up to 26 phases
Adjacent voxel, the voxel only intersected with growing element inner plane just have probability to be merged.Assuming that the plane equation in growing element
For ax+by+cz+d=0；A0-A7(xi,yi,zi) represent 8 vertex of an adjacent voxels.If 8 vertex meet following
Any inequality then indicates that vertex all in the side of plane, indicates that plane and voxel are non-intersecting；Each growing element is only recorded
The adjacent voxels of intersection establish neighborhood index.
axi+byi+czi+d≤0
axi+byi+czi+d≥0
Step 2, it is indexed according to the extracted growing element of step 1 and the neighborhood of foundation, carries out area growth process, it is real
The detailed process that example is implemented is described as follows:
Step 2.1, the growing element obtained from step 1 choose not processed voxel comprising maximum number of points as kind
Sub- plane is added in growth queue Q, and labeled as processed；
Step 2.2 establishes region growing criterion.Criterion of Selecting in this example: the angle of 1 plane and seed plane is less than
Threshold θ2(15 ° of value in this example), the distance between 2 planes and seed plane are less than threshold value d2(this example value 0.12m)；
If step 2.3, growth queue are not empty, one voxel of taking-up from growth queue Q.It is established using step 1
Index, judges whether the not processed neighboring voxels of the voxel meet growth criterion, if meeting criterion, which is closed
And it is concentrated to seed point, while the voxel is added to growth queue, and labeled as processed.The step repeats n times, Zhi Daosheng
Until queue is classified as sky, then a complete initial plane is detected.
Step 2.3, step 2.1-2.3 are repeated as many times, until all voxels are all processed.Obtain initial planar set
It closes, the sample result is as shown in Figure 4.
Step 3, since the degree of roughness and scale on rock mass surface are different, so according to the extracted initial plane of step 2
Often contain more holiday, it is therefore desirable to grow to left point cloud using realm information, the tool that example is implemented
Body procedure declaration is as follows:
Step 3.1 chooses not processed putting down comprising maximum number of points from the extracted initial plane set of step 2
Then the point of this internal plane and neighborhood is chosen as growth candidate point cloud R in face from left point cloud
Step 3.2 establishes growth criterion: the angle of 1 point and objective plane is less than threshold θ3(20 ° of value in this example), 2
The distance between point and objective plane are less than threshold value d3(this example value 0.2m)；
Step 3.3 judges that each point in left point cloud R, the point for meeting growth criterion is added to objective plane,
Obtain final plane point set；
Step 3.4, step 3.1-3.3 are repeated several times, and until all initial planes are all processed, are obtained final flat
Millet cake cloud.Final result is as shown in Figure 5 in this example.
The technological invention, which compares traditional characteristic face extracting method in terms of operation efficiency and plane extract accuracy rate, to be had greatly
It improves, for the example method, by quantitative analysis, its accuracy has reached 90.35%, and time loss is in 2.473s, accurately
Degree and efficiency have significant raising compared to conventional method, can be used for rock mass numerical analysis and subsequent three-dimensional rock mass modeling
Process.
Specific embodiment only discussed herein above is only to be done to spirit of that invention for example, it is noted that, ability
Field technique personnel described specific embodiment can be made under the premise of not departing from inventive principle it is any modification, variation or
It is equivalent, but these modifications, variation and equivalent will fall within the scope of protection of the present invention.
Claims (4)
1. a kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel, it is characterised in that mainly forgive with
Lower step:
Step 1, voxel-based space are divided to be detected with coplanarity: the normal vector of each point in first choice estimation point cloud；Then root
Voxel is quickly established according to the bounding box size of rock mass point cloud；Coplanarity is carried out in each voxel inner utilization stochastical sampling consistency
Detection is to establish growing element (plane)；Neighborhood index finally is established to each growing element.
Step 2, the rock mass point cloud initial plane based on algorithm of region growing are extracted: according to the growing element after foundation, utilizing base
Initial plane set is extracted in the algorithm of region growing of voxel；
Step 3, the planar growth based on neighborhood relationships: on the basis of each plane after extracting, establishing point cloud merging criterion,
The left point cloud after extraction is determined in the neighborhood of its plane, obtains final plane set.
2. according to right want 1 described in a kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel,
It is characterized by: the step 1 further includes steps of
Step 1.1 utilizes the normal vector of each point in principal component analysis (PCA) estimation point cloud.For any point p in cloud,
It in total include k point in the region that its radius is r.Then this k point constructs covariance matrix Σ, wherein λ1< λ2< λ3It indicates
Three characteristic values of covariance matrix Σ, v1,v2,v3For its corresponding feature vector.Then its minimal eigenvalue λ1Corresponding feature
Vector v1For the normal vector of p point.
Σ·vj=λj·vj,j∈{1,2,3}
The given voxel side length of step 1.2, basis, will put cloud rapid voxel.
Step 1.3, for each voxel inside, if count be greater than certain threshold value (N_r) if utilize based on normal vector with
Machine samples consistency (ND-RANSAC) and carries out coplanarity detection, for extracting growing element (plane).It is assumed that inside each voxel
Containing N number of point, therefrom randomly selects 3 non-coplanar points and establishes plane pl, given a mark according to remaining cloud to this plane,
If putting the threshold value d being less than to the distance of plane1And the normal direction of point and the angle of plane normal direction are less than threshold θ1Then score adds 1；This mistake
Journey iteration T times, the maximum plane of score are final growing element (plane)；
Step 1.4, according to the growing element of foundation on the basis of, to each growing element establish neighborhood index.
3. according to right want 1 described in a kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel,
It is characterized by: the step 2 further includes steps of
Step 2.1 therefrom chooses comprising the most plane of point number as seed plane the growing element that step 1 is extracted；
Step 2.2 establishes region growing criterion: the normal angle of 1 liang of plane is less than certain threshold θ2；The distance of 2 liang of planes is small
In d2。
Step 2.3, centered on seed plane, indexed using the established neighborhood of step 1.4, its peripheral neighborhood searched
Rope；The growing element for meeting growth criterion is merged；The unit being newly added continues to grow as new seed plane,
Until not new plane addition, then a complete plane is extracted from cloud.
Step 2.4, the process iteration of step 2.1-2.3 carry out for several times, until all growing elements are all processed, obtain just
Beginning plane set.
4. according to right want 1 described in a kind of region growing 3D rock mass point cloud plane extracting method based on high quality voxel,
It is characterized by: the step 3 further includes steps of
Step 3.1, the initial plane set extracted every time from step 2 choose the maximum plane P that counts；
Step 3.2, from the left point cloud after extraction, select P neighborhood and inside left point converge close R；
Step 3.3 establishes point cloud merging criterion: if 1 point is less than certain threshold θ apart from normal angle to plane P3；2 points are arrived
Plane P distance is less than d3.Point each in R is determined, the point for meeting merging criterion is added in plane P, and is estimated again
Calculate plane parameter；
Step 3.4, repeat the above steps for each initial plane that step 2 is extracted 3.1-3.4, obtains final planar set
It closes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811603927.9A CN109685821A (en) | 2018-12-26 | 2018-12-26 | Region growing 3D rock mass point cloud plane extracting method based on high quality voxel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811603927.9A CN109685821A (en) | 2018-12-26 | 2018-12-26 | Region growing 3D rock mass point cloud plane extracting method based on high quality voxel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109685821A true CN109685821A (en) | 2019-04-26 |
Family
ID=66189776
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811603927.9A Pending CN109685821A (en) | 2018-12-26 | 2018-12-26 | Region growing 3D rock mass point cloud plane extracting method based on high quality voxel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109685821A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458854A (en) * | 2018-05-02 | 2019-11-15 | 北京图森未来科技有限公司 | A kind of road edge detection method and device |
| CN111178214A (en) * | 2019-12-23 | 2020-05-19 | 天津大学 | High steep slope dangerous rock mass rapid identification method based on unmanned aerial vehicle photography technology |
| CN111985507A (en) * | 2020-08-28 | 2020-11-24 | 东北大学 | Rock three-dimensional point cloud joint trace extraction method |
| CN112597796A (en) * | 2020-11-18 | 2021-04-02 | 中国石油大学(华东) | Robust point cloud representation learning method based on graph convolution |
| CN116071530A (en) * | 2023-03-07 | 2023-05-05 | 青岛市勘察测绘研究院 | Building roof voxelized segmentation method based on airborne laser point cloud |
| CN116482711A (en) * | 2023-06-21 | 2023-07-25 | 之江实验室 | Local static environment sensing method and device for autonomous selection of landing zone |
| CN117689965A (en) * | 2024-02-04 | 2024-03-12 | 山东黄金矿业股份有限公司新城金矿 | Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm |
| CN117745728A (en) * | 2024-02-21 | 2024-03-22 | 法奥意威(苏州)机器人系统有限公司 | Point cloud plane detection method and device, electronic equipment and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150199839A1 (en) * | 2012-08-02 | 2015-07-16 | Earthmine, Inc. | Three-Dimentional Plane Panorama Creation Through Hough-Based Line Detection |
| CN106896213A (en) * | 2017-02-22 | 2017-06-27 | 中国地质大学（武汉） | A kind of rock mass discontinuity Intelligent Recognition and information extracting method based on cloud data |
| CN107657659A (en) * | 2017-08-14 | 2018-02-02 | 南京航空航天大学 | The Manhattan construction method for automatic modeling of scanning three-dimensional point cloud is fitted based on cuboid |
| CN107945189A (en) * | 2017-11-28 | 2018-04-20 | 武汉大学 | A kind of point cloud plane dividing method based on normal distribution transform unit |
-
2018
- 2018-12-26 CN CN201811603927.9A patent/CN109685821A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150199839A1 (en) * | 2012-08-02 | 2015-07-16 | Earthmine, Inc. | Three-Dimentional Plane Panorama Creation Through Hough-Based Line Detection |
| CN106896213A (en) * | 2017-02-22 | 2017-06-27 | 中国地质大学（武汉） | A kind of rock mass discontinuity Intelligent Recognition and information extracting method based on cloud data |
| CN107657659A (en) * | 2017-08-14 | 2018-02-02 | 南京航空航天大学 | The Manhattan construction method for automatic modeling of scanning three-dimensional point cloud is fitted based on cuboid |
| CN107945189A (en) * | 2017-11-28 | 2018-04-20 | 武汉大学 | A kind of point cloud plane dividing method based on normal distribution transform unit |
Non-Patent Citations (2)
| Title |
|---|
| 李明磊等: "采用八叉树体素生长的点云平面提取", 《光学精密工程》 * |
| 肖博文: "基于平面提取的三维点云配准算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458854A (en) * | 2018-05-02 | 2019-11-15 | 北京图森未来科技有限公司 | A kind of road edge detection method and device |
| CN110458854B (en) * | 2018-05-02 | 2022-11-15 | 北京图森未来科技有限公司 | Road edge detection method and device |
| US11500101B2 (en) | 2018-05-02 | 2022-11-15 | Tusimple, Inc. | Curb detection by analysis of reflection images |
| CN111178214A (en) * | 2019-12-23 | 2020-05-19 | 天津大学 | High steep slope dangerous rock mass rapid identification method based on unmanned aerial vehicle photography technology |
| CN111985507A (en) * | 2020-08-28 | 2020-11-24 | 东北大学 | Rock three-dimensional point cloud joint trace extraction method |
| CN111985507B (en) * | 2020-08-28 | 2023-07-28 | 东北大学 | Rock mass three-dimensional point cloud joint trace extraction method |
| CN112597796A (en) * | 2020-11-18 | 2021-04-02 | 中国石油大学(华东) | Robust point cloud representation learning method based on graph convolution |
| CN116071530A (en) * | 2023-03-07 | 2023-05-05 | 青岛市勘察测绘研究院 | Building roof voxelized segmentation method based on airborne laser point cloud |
| CN116482711A (en) * | 2023-06-21 | 2023-07-25 | 之江实验室 | Local static environment sensing method and device for autonomous selection of landing zone |
| CN117689965A (en) * | 2024-02-04 | 2024-03-12 | 山东黄金矿业股份有限公司新城金矿 | Rapid identification method for laser point cloud rock mass structure based on RANSAC algorithm |
| CN117745728A (en) * | 2024-02-21 | 2024-03-22 | 法奥意威(苏州)机器人系统有限公司 | Point cloud plane detection method and device, electronic equipment and readable storage medium |
| CN117745728B (en) * | 2024-02-21 | 2024-05-24 | 法奥意威(苏州)机器人系统有限公司 | Point cloud plane detection method and device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109685821A (en) | Region growing 3D rock mass point cloud plane extracting method based on high quality voxel | |
| CN112070769B (en) | Layered point cloud segmentation method based on DBSCAN | |
| Ge et al. | Enhanced subpixel mapping with spatial distribution patterns of geographical objects | |
| CN109034065B (en) | Indoor scene object extraction method based on point cloud | |
| CN114492619B (en) | Point cloud data set construction method and device based on statistics and concave-convex performance | |
| CN113255677B (en) | Method, equipment and medium for rapidly extracting rock mass structural plane and occurrence information | |
| CN107945189A (en) | A kind of point cloud plane dividing method based on normal distribution transform unit | |
| CN113963130B (en) | Construction method of fracture network model for rock core fracture | |
| CN113409332B (en) | Building plane segmentation method based on three-dimensional point cloud | |
| CN115546116A (en) | Method and system for extracting and calculating spacing of discontinuous surface of fully-covered rock mass | |
| CN111161267A (en) | Segmentation method of three-dimensional point cloud model | |
| CN107481243B (en) | Sheep body size detection method based on sheep top view | |
| CN110348478B (en) | Method for extracting trees in outdoor point cloud scene based on shape classification and combination | |
| Liu et al. | Application of three-dimensional laser scanning in the protection of multi-dynasty ceramic fragments | |
| CN107993242B (en) | Method for extracting boundary of missing area based on airborne LiDAR point cloud data | |
| CN112581511B (en) | Three-dimensional reconstruction method and system based on near vertical scanning point cloud rapid registration | |
| CN112561989B (en) | Recognition method for hoisting object in construction scene | |
| CN109035311A (en) | A kind of curved bone fracture autoregistration and internal fixation steel plate pre-bending modeling method | |
| He et al. | A curvature based adaptive neighborhood for individual point cloud classification | |
| Ni et al. | Applications of 3d-edge detection for als point cloud | |
| Tan et al. | Semantic enrichment for rooftop modeling using aerial LiDAR reflectance | |
| CN116452604A (en) | Complex substation scene segmentation method, device and storage medium | |
| CN117253205A (en) | Road surface point cloud rapid extraction method based on mobile measurement system | |
| CN116010644A (en) | Axial movement and local matching-based repetitive building element retrieval method | |
| CN114429537A (en) | Method for extracting niche from grotto point cloud |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190426 |
|
| WD01 | Invention patent application deemed withdrawn after publication |