CN114049387A - Tree skeleton extraction method based on 3D point cloud - Google Patents
Tree skeleton extraction method based on 3D point cloud Download PDFInfo
- Publication number
- CN114049387A CN114049387A CN202111231714.XA CN202111231714A CN114049387A CN 114049387 A CN114049387 A CN 114049387A CN 202111231714 A CN202111231714 A CN 202111231714A CN 114049387 A CN114049387 A CN 114049387A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- cloud data
- skeleton
- tree
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Abstract
The invention discloses a tree skeleton extraction method based on 3D point cloud, and relates to the technical field of big data processing and tree 3D modeling. The invention comprises the following steps: 3D point cloud data scale is reduced by using an octree and L1-median point method; constructing a ground root node; constructing a K-nearest neighbor graph and dividing a level hierarchical set; clustering is carried out to obtain point cloud data of different branches in the same horizontal level set; dividing by using an octree method, and calculating an L1-median point of each division to obtain skeleton points; according to whether the K-neighbors of the members in the two clusters have intersection, whether the two clusters are adjacent is defined, and framework points of the adjacent clusters are connected to obtain an initial framework: and calculating a B-spline curve of the initial skeleton to obtain an optimized skeleton. According to the method, the influence of the uneven distribution of the 3D point cloud on the skeleton extraction can be effectively reduced, so that the extracted tree skeleton result is more accurate.
Description
Technical Field
The invention belongs to the technical field of big data processing and tree 3D modeling, and particularly relates to a tree skeleton extraction method based on 3D point cloud.
Background
The skeleton is used as a 'compact' representation method of the three-dimensional model, and the topological structure and the posture characteristics of the plant model can be abstractly reflected. Because the skeleton has simple topological structure and convenient operation, and is a basic element for surface reconstruction, retrieval and matching of the three-dimensional model, the skeleton information is widely applied to the fields of plant modeling, three-dimensional animation design, medical images, mapping images and the like.
Tree modeling has wide application in computational forestry, functional structural plant modeling, city planning, and many other fields. The tree skeleton is the basis for reconstructing a biological model and a structural model of the tree, and has important significance in tree three-dimensional modeling and forestry research. At present, many tree skeleton extraction methods have been proposed. Verroust and Lazarus propose a horizontal setting method, which can extract the skeleton of the tree-like object. They use geodesic distances on the K-nearest neighbor graph to generate multiple level sets. The method can capture the basic topology of the tree, is easy to realize, and is widely used. However, this method does not take into account the varying density or missing data in the point cloud and does not guarantee the centering of the skeleton. Cao et al established a simple K-nearest neighbor-based point cloud topological relationship, and attempted to generalize the Laplacian shrinkage method to the point cloud model skeleton extraction. The method requires that the input point cloud is dense enough and free of noise, but due to the complex plant morphological structure, serious self-shielding and the precision limitation of common acquisition equipment, a large amount of noise exists in the acquired plant initial cloud data, and the loss of local data is serious, so that the skeleton obtained by the method is not accurate in the actual situation. Livny et al propose an algorithm for reconstructing a plant point cloud skeleton, which defines a skeleton structure as a branch structure diagram (BSG), and adopts Dijkstra algorithm to construct a minimum weight path diagram by taking a weight as a path according to a root node as a starting point and a square of an Euclidean distance between any two points as a weight. However, in this method, the selection of the root node is very important, and if the root node is not properly selected, the difference between the structure of the skeleton near the root node of the skeleton and the actual structure is very different. A 3D plant skeleton extraction and adjustment method is proposed in patent literature (patent application publication No. CN 109816775 a). The method comprises the steps of clustering plant point cloud models, and then, mainly extracting and adjusting skeletons aiming at plant leaves. However, this patent ignores the special structural features of the trunk portion of the tree.
The 3D point cloud data for tree modeling has the characteristics of large scale and containing a large amount of noise data, the existing tree skeleton extraction method is easily influenced by noise in the point cloud data and improper selection of root nodes, and the extracted skeleton cannot ensure centering performance and accuracy. Therefore, the method for rapidly and accurately extracting the tree skeleton, which is suitable for large-scale point cloud data, has important practical significance.
Disclosure of Invention
The invention aims to provide a tree skeleton extraction method based on 3D point cloud, which solves the existing problems: the tree skeleton extraction method is easily affected by noise in the point cloud data and improper selection of root nodes, and the extracted skeleton cannot ensure centering performance and accuracy.
In order to solve the technical problems, the invention is realized by the following technical scheme: a tree skeleton extraction method based on 3D point cloud comprises the following steps:
reducing the scale of the 3D point cloud data by using an octree and L1-median point method to obtain a reduced point cloud data set;
constructing a ground root node, and adding the node into the reduced point cloud data set;
constructing a K-nearest neighbor graph for the reduced point cloud data set added with the ground root node, calculating the path distance from all other points to the origin point on the K-nearest neighbor graph by taking the ground root node as the origin point, and setting the range of the path distance of each horizontal level set, thereby dividing the point cloud data into the horizontal level sets;
clustering the point cloud data in each horizontal level set by a method of searching a connected subgraph of a beta-neighborhood graph, thereby obtaining point cloud data of different branches in the same horizontal level set;
repeatedly dividing each cluster clustered in the steps by using an octree method, calculating an L1-median point of each division until a new point cloud data set only contains one point cloud data, wherein the point cloud data is a skeleton point of the cluster, and repeating the steps until all clusters in all horizontal hierarchies are traversed;
and according to whether the K-neighbors of the members in the two clusters have intersection, defining whether the two clusters are adjacent, connecting the skeleton points of the adjacent clusters to obtain an initial skeleton, and calculating a B-spline curve of the initial skeleton to obtain an optimized skeleton.
Further: the method for reducing the 3D point cloud data scale by using the octree and the L1-median point is used for obtaining a reduced point cloud data set, and the method mainly comprises the following steps:
for an original 3D point cloud data setXDividing recursively by using octree method, and collecting point cloud data in a certain divisionPWhen the number of points is less than 20, stopping the pairPThe division of (a) into (b),
for each partitionPCalculate its L1-median point asM；
Will be provided withMAdding to a new point cloud datasetX’In, a new point cloud data setX’I.e. the reduced point cloud data set.
Further: wherein, construct a ground root node, and add this point into the point cloud data set after said reduction, mainly include:
selecting all points from the lowest horizontal plane to the upper direction of the lowest horizontal plane of the longitudinal axis within the total length 1/15 of the longitudinal axis as a point cloud data set of the root of the treeR；
Recursively applying octree method to point cloud data set of tree rootRDividing, stopping recursion when the point cloud number in the division result is less than 5, and calculating the L1-median point of each division to obtain new pointsCloud collectionR’；
Recalculating a new set of point cloudsR’L1-median point, modifying the longitudinal axis coordinate of the point to the position below the root, and obtaining the reconstructed ground root noder；
Constructing the obtained ground root noderAdding to the reduced point cloud datasetX’。
Further: the method includes the steps that a K-nearest neighbor graph is constructed for a point cloud data set after reduction after ground root nodes are added, path distances from all other points to an original point are calculated on the K-nearest neighbor graph by taking the ground root nodes as the original point, and the range of the path distance of each horizontal level set is set, so that the point cloud data are divided into the horizontal level sets, and the method mainly comprises the following steps:
reduced point cloud data setX’Constructing a K-nearest neighbor graph;
computing each other point on a K-nearest neighbor graphiTo the ground root noderShortest path distance ofd s i；
Is provided with the firstmThe specific method of the path distance range of the individual horizontal hierarchy set is as follows: first, the maximum shortest path distance is obtainedmaxd s =max(d s i) Divide it intoL+1 layers, each layer having a length ofunitL=maxd s / L Of 1 atmThe distance range of the path from the horizontal hierarchy set to the ground root node is (m-1)*unitL~m* unitL Wherein 1 is less than or equal tom≤L+1；
Dividing the set of belonged levels according to the shortest path distance from each point to the ground root node, specifically, aiming at the pointiCalculate its affiliated levelLevelI.e. byLevel=d s i%unitL+1, then pointiIs divided intoLevelAnd (3) a layer.
Further: the method for searching the connected subgraph of the beta-neighborhood graph is used for clustering the point cloud data in each horizontal level set, so that point cloud data of different branches in the same horizontal level set are obtained, and the method mainly comprises the following steps:
for eachPoints included in a horizontal hierarchical set, each point calculatediMean of distances to its K-nearest neighborsd m iCalculate alld m iMean value ofMAnd standard deviation ofSSetting a constraint valueβ=M-nSWherein, 0 is less than or equal ton≤2
For each horizontal hierarchical set, connecting each point with all points in the beta neighborhood range thereof to form a beta-neighborhood graph, clustering each horizontal hierarchical set by searching connected subgraphs in the beta-neighborhood graph,
defining non-clustered points as noise points from the point cloud datasetX’Is deleted.
Further: the method comprises the following steps of defining whether two clusters are adjacent or not according to whether the K-neighbors of members in the two clusters have intersection or not, connecting framework points of the adjacent clusters to obtain an initial framework, calculating a B-spline curve of the initial framework to obtain an optimized framework, and mainly comprising the following steps of:
for each class clusterCThe union of the K-neighbors of all the points in the cluster is obtained asCThe extended cluster of (3);
for any two cluster classesC i AndC j determining whether the two clusters are adjacent according to whether the intersection of the extended clusters is empty, and storing adjacent information into an adjacent matrix;
arbitrarily selecting skeleton points of two adjacent clustersAAndBconnecting, and repeating the steps until all the adjacent clusters are traversed to obtain an initial skeleton;
and calculating a B-spline curve of the initial skeleton to obtain the optimized tree skeleton.
The invention has the following beneficial effects:
the method for extracting the 3D tree point cloud model skeleton can effectively reduce the influence of the 3D point cloud distribution nonuniformity on skeleton extraction, so that the extracted tree skeleton result is more accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of the extraction method of the present invention.
FIG. 2 is a flow chart of the method for reducing the scale of 3D point cloud data according to the present invention.
Fig. 3 is a flowchart of a method for constructing a ground root node according to the present invention.
FIG. 4 is a flow chart of a method for partitioning a horizontal hierarchical set in accordance with the present invention.
FIG. 5 is a flow chart of the method for dividing different branches according to the present invention.
FIG. 6 is a flow chart of a skeleton point calculation method according to the present invention.
FIG. 7 is a flow chart of the method for connecting and optimizing the skeleton points according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Specifically, referring to fig. 1, fig. 1 is a flowchart of a tree skeleton extraction method based on 3D point cloud.
The tree skeleton extraction method based on the 3D point cloud comprises the following steps:
step 1:
and 3D point cloud data size is reduced by using an octree and L1-median point method.
Referring to fig. 2, in an embodiment, the step 1 includes the following steps:
step 1.1: for an original 3D point cloud data setXDividing recursively by using octree method, and collecting point cloud data in a certain divisionPWhen the number of points is less than 20, stopping the pairPThe division of (a) into (b),
step 1.2 for each partitionPCalculate its L1-median point asM；
Step 1.3: mixingMAdding to a new point cloud datasetX’In, a new point cloud data setX’I.e. reduced point cloud data set
Step 2:
and constructing a ground root node, and adding the node into the reduced tree point cloud data set.
Referring to fig. 3, in an embodiment, the step 2 includes the following steps:
step 2.1: selecting all points from the lowest horizontal plane to the upper direction of the lowest horizontal plane of the longitudinal axis within the total length 1/15 of the longitudinal axis as a point cloud data set of the root of the treeR；
Step 2.2: recursively applying octree method to point cloud data set of tree rootRDividing, stopping recursion when the point cloud number in the division result is less than 5, and calculating the L1-median point of each division to obtain a new point cloud setR’；
Step 2.3: recalculating a new set of point cloudsR’L1-median point, modifying the longitudinal axis coordinate of the point to the position below the root, and obtaining the reconstructed ground root noder；
Step 2.4: constructing the obtained ground root noderAdding to the reduced point cloud datasetX’。
And step 3:
for the obtained new point cloud setX’And constructing a K-nearest neighbor graph, calculating the path distance from all other points to the origin point on the K-nearest neighbor graph by taking the ground root node as the origin point, and setting the range of the path distance of each horizontal level set, thereby dividing the point cloud data into the horizontal level sets.
Referring to fig. 4, the method for dividing the level hierarchy set in step 3 includes the following steps:
3.1: constructing a K-nearest neighbor graph on the point cloud data set X';
3.2 calculate every other Point on the K-nearest neighbors graphiTo the ground root noderShortest path distance ofd s i；
3.3 setting upmThe specific method of the path distance range of the individual horizontal hierarchy set is as follows: first, the maximum shortest path distance is obtainedmaxd s =max(d s i) Divide it intoL+1 layers, each layer having a length ofunitL=maxd s / L Of 1 atmThe distance range of the path from the horizontal hierarchy set to the ground root node is (m-1)*unitL~m* unitL Wherein 1 is less than or equal tom≤L+1；
3.4 partitioning the set of belonged levels according to the shortest path distance from each point to the ground root node, in particular, for a pointiCalculate its affiliated levelLevelI.e. byLevel=d s i%unitL+1, then pointiIs divided intoLevelAnd (3) a layer.
And 4, step 4:
and clustering the point cloud data in each horizontal hierarchical set by a method of searching a connected subgraph of the beta-neighborhood graph, thereby obtaining the point cloud data of different branches in the same horizontal hierarchical set.
Specifically referring to fig. 5, the step 4 of obtaining point cloud data of different branches in the same horizontal level set includes the following steps:
step 4.1: calculating each point for the points contained in each horizontal hierarchical setiMean of distances to its K-nearest neighborsd m iCalculate alld m iMean value ofMAnd standard deviation ofSSetting a constraint valueβ=M-nSWherein, 0 is less than or equal ton≤2;
Step 4.2: connecting each point with all points in the beta neighborhood range of each horizontal hierarchical set so as to form a beta-neighborhood graph, and clustering each horizontal hierarchical set by searching a connected subgraph in the beta-neighborhood graph;
step 4.3: defining non-clustered points as noise points from the point cloud datasetX’Is deleted.
And 5:
and (4) repeatedly using an octree method to divide each cluster in the step (4), and calculating L1-median points of each division until a new point cloud data set only contains one point cloud data, wherein the point cloud data is the skeleton point of the cluster.
Specifically referring to fig. 6, the skeleton point calculation method in step 5 includes the following steps:
step 5.1: repeatedly dividing the point cloud data sets of different clusters in the same horizontal level set by using an octree method, and calculating an L1-median point of each division until a new point cloud data set only contains one point cloud data, wherein the point cloud data is a skeleton point of the cluster;
step 5.2: and repeating the step 5.1 until all the class clusters in all the horizontal hierarchical sets are traversed.
Step 6:
and defining whether the two class clusters are adjacent or not according to whether the K-neighbors of the members in the two class clusters have intersection or not. And then connecting the skeleton points of the adjacent clusters to obtain an initial skeleton, and calculating a B-spline curve of the initial skeleton to obtain an optimized skeleton.
Referring to fig. 7, the method for connecting and optimizing skeleton points in step 6 includes the following steps:
step 6.1: for each class clusterCThe union of the K-neighbors of all the points in the cluster is obtained asCThe extended cluster of (3);
step 6.2, aiming at any two clustersC i AndC j determining whether the two clusters are adjacent according to whether the intersection of the extended clusters is empty, and storing adjacent information into an adjacent matrix;
step 6.3: arbitrarily selecting skeleton points of two adjacent clustersAAndBconnecting, and repeating the steps until all the adjacent clusters are traversed to obtain an initial skeleton;
step 6.4: and calculating a B-spline curve of the initial skeleton to obtain the optimized tree skeleton.
The method is integrated as follows: the method for extracting the 3D tree point cloud model skeleton can effectively reduce the influence of the 3D point cloud distribution nonuniformity on skeleton extraction, so that the extracted tree skeleton result is more accurate.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (6)
1. A tree skeleton extraction method based on 3D point cloud is characterized by comprising the following steps: the method comprises the following steps:
reducing the scale of the 3D point cloud data by using an octree and L1-median point method to obtain a reduced point cloud data set;
constructing a ground root node, and adding the node into the reduced point cloud data set;
constructing a K-nearest neighbor graph for the reduced point cloud data set added with the ground root node, calculating the path distance from all other points to the origin point on the K-nearest neighbor graph by taking the ground root node as the origin point, and setting the range of the path distance of each horizontal level set, thereby dividing the point cloud data into the horizontal level sets;
clustering the point cloud data in each horizontal level set by a method of searching a connected subgraph of a beta-neighborhood graph, thereby obtaining point cloud data of different branches in the same horizontal level set;
repeatedly dividing each cluster clustered in the steps by using an octree method, calculating an L1-median point of each division until a new point cloud data set only contains one point cloud data, wherein the point cloud data is a skeleton point of the cluster, and repeating the steps until all clusters in all horizontal hierarchies are traversed;
and according to whether the K-neighbors of the members in the two clusters have intersection, defining whether the two clusters are adjacent, connecting the skeleton points of the adjacent clusters to obtain an initial skeleton, and calculating a B-spline curve of the initial skeleton to obtain an optimized skeleton.
2. The tree skeleton extraction method based on the 3D point cloud as claimed in claim 1, wherein the tree skeleton extraction method comprises the following steps: the method for reducing the 3D point cloud data scale by using the octree and the L1-median point is used for obtaining a reduced point cloud data set, and the method mainly comprises the following steps:
for an original 3D point cloud data setXDividing recursively by using octree method, and collecting point cloud data in a certain divisionPWhen the number of points is less than 20, stopping the pairPThe division of (a) into (b),
for each partitionPCalculate its L1-median point asM；
Will be provided withMAdding to a new point cloud datasetX’In, a new point cloud data setX’I.e. the reduced point cloud data set.
3. The tree skeleton extraction method based on the 3D point cloud as claimed in claim 1, wherein the tree skeleton extraction method comprises the following steps: wherein, construct a ground root node, and add this point into the point cloud data set after said reduction, mainly include:
selecting all points from the lowest horizontal plane to the upper direction of the lowest horizontal plane of the longitudinal axis within the total length 1/15 of the longitudinal axis as a point cloud data set of the root of the treeR；
Recursively applying octree method to point cloud data set of tree rootRDividing, stopping recursion when the point cloud number in the division result is less than 5, and calculating the L1-median point of each division to obtain a new point cloud setR’；
Recalculating a new set of point cloudsR’L1-median point, modifying the longitudinal axis coordinate of the point to the position below the root, and obtaining the reconstructed ground root noder；
Constructing the obtained ground root noderAdding to the reduced point cloud datasetX’。
4. The tree skeleton extraction method based on the 3D point cloud as claimed in claim 1, wherein the tree skeleton extraction method comprises the following steps: the method includes the steps that a K-nearest neighbor graph is constructed for a point cloud data set after reduction after ground root nodes are added, path distances from all other points to an original point are calculated on the K-nearest neighbor graph by taking the ground root nodes as the original point, and the range of the path distance of each horizontal level set is set, so that the point cloud data are divided into the horizontal level sets, and the method mainly comprises the following steps:
reduced point cloud data setX’Constructing a K-nearest neighbor graph;
computing each other point on a K-nearest neighbor graphiTo the ground root noderShortest path distance ofd s i；
Is provided with the firstmThe specific method of the path distance range of the individual horizontal hierarchy set is as follows: first, the maximum shortest path distance is obtainedmaxd s =max(d s i) Divide it intoL+1 layers, each layer having a length ofunitL=maxd s /LOf 1 atmThe distance range of the path from the horizontal hierarchy set to the ground root node is (m-1)*unitL~m*unitLWherein 1 is less than or equal tom≤L+1；
Dividing the set of belonged levels according to the shortest path distance from each point to the ground root node, specifically, aiming at the pointiCalculate its affiliated levelLevelI.e. byLevel=d s i%unitL+1, then pointiIs divided intoLevelAnd (3) a layer.
5. The tree skeleton extraction method based on the 3D point cloud as claimed in claim 1, wherein the tree skeleton extraction method comprises the following steps: the method for searching the connected subgraph of the beta-neighborhood graph is used for clustering the point cloud data in each horizontal level set, so that point cloud data of different branches in the same horizontal level set are obtained, and the method mainly comprises the following steps:
calculating each point for the points contained in each horizontal hierarchical setiMean of distances to its K-nearest neighborsd m iCalculate alld m iMean value ofMAnd standard deviation ofSSetting a constraint valueβ=M-nSWherein, 0 is less than or equal ton≤2
For each horizontal hierarchical set, connecting each point with all points in the beta neighborhood range thereof to form a beta-neighborhood graph, clustering each horizontal hierarchical set by searching connected subgraphs in the beta-neighborhood graph,
defining non-clustered points as noise points from the point cloud datasetX’Is deleted.
6. The tree skeleton extraction method based on the 3D point cloud as claimed in claim 1, wherein the tree skeleton extraction method comprises the following steps: the method comprises the following steps of defining whether two clusters are adjacent or not according to whether the K-neighbors of members in the two clusters have intersection or not, connecting framework points of the adjacent clusters to obtain an initial framework, calculating a B-spline curve of the initial framework to obtain an optimized framework, and mainly comprising the following steps of:
for each class clusterCThe union of the K-neighbors of all the points in the cluster is obtained asCThe extended cluster of (3);
for any two cluster classesC i AndC j determining whether the two clusters are adjacent according to whether the intersection of the extended clusters is empty, and storing adjacent information into an adjacent matrix;
arbitrarily selecting skeleton points of two adjacent clustersAAndBconnecting, and repeating the steps until all the adjacent clusters are traversed to obtain an initial skeleton;
and calculating a B-spline curve of the initial skeleton to obtain the optimized tree skeleton.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111231714.XA CN114049387A (en) | 2021-10-22 | 2021-10-22 | Tree skeleton extraction method based on 3D point cloud |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111231714.XA CN114049387A (en) | 2021-10-22 | 2021-10-22 | Tree skeleton extraction method based on 3D point cloud |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114049387A true CN114049387A (en) | 2022-02-15 |
Family
ID=80205671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111231714.XA Pending CN114049387A (en) | 2021-10-22 | 2021-10-22 | Tree skeleton extraction method based on 3D point cloud |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114049387A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311418A (en) * | 2022-10-10 | 2022-11-08 | 深圳大学 | Multi-detail-level tree model single reconstruction method and device |
-
2021
- 2021-10-22 CN CN202111231714.XA patent/CN114049387A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311418A (en) * | 2022-10-10 | 2022-11-08 | 深圳大学 | Multi-detail-level tree model single reconstruction method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Verdie et al. | LOD generation for urban scenes | |
| CN105825550B (en) | Take the complex three-dimensional building model cutting modeling method of consistency into account | |
| Tsai | Delaunay triangulations in TIN creation: an overview and a linear-time algorithm | |
| CN105139379B (en) | Based on the progressive extracting method of classified and layered airborne Lidar points cloud building top surface | |
| CN104462163B (en) | A kind of threedimensional model characterizing method, search method and searching system | |
| CN106780458B (en) | Point cloud framework extraction method and device | |
| CN106095907A (en) | Based on the laser point cloud data management method that Octree is integrated with three-dimensional R star tree | |
| CN108765568A (en) | A kind of multi-level building quick three-dimensional reconstructing method based on laser radar point cloud | |
| CN104731984B (en) | Automobile wheel hub surface sampling point R tree overflow node incremental clustering optimization method | |
| WO2023124676A1 (en) | 3d model construction method, apparatus, and electronic device | |
| CN113468286A (en) | Three-dimensional pyramid construction method based on triangular patch number division | |
| CN114049387A (en) | Tree skeleton extraction method based on 3D point cloud | |
| CN107818338A (en) | A kind of method and system of building group pattern-recognition towards Map Generalization | |
| CN106373192A (en) | Non-topology-consistency three-dimensional grid block tracking algorithm | |
| CN110738726B (en) | Robot vision-guided three-dimensional object reconstruction method based on octree | |
| CN117034051A (en) | Water conservancy information aggregation method, device and medium based on BIRCH algorithm | |
| Breunig et al. | 3D geo-DBMS | |
| CN116258840A (en) | Hierarchical detail representation tree generation method, device, equipment and storage medium | |
| CN116152451A (en) | Multidimensional parameterized city information model construction method, system and computer equipment | |
| CN104881498B (en) | The Out-of-Core Quick uniform compressing methods of massive point cloud | |
| CN114332322A (en) | Big data rendering method of monomer model in three-dimensional scene | |
| WO2021232442A1 (en) | Density clustering method and apparatus on basis of dynamic grid hash index | |
| Mascarenhas et al. | Isocontour based visualization of time-varying scalar fields | |
| CN111915720A (en) | Automatic conversion method from building Mesh model to CityGML model | |
| CN113342999B (en) | Variable-resolution-ratio point cloud simplification method based on multi-layer skip sequence tree structure |
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 |