CN111598915B - Point cloud single wood segmentation method, device, equipment and computer readable medium - Google Patents
Point cloud single wood segmentation method, device, equipment and computer readable medium Download PDFInfo
- Publication number
- CN111598915B CN111598915B CN202010422711.3A CN202010422711A CN111598915B CN 111598915 B CN111598915 B CN 111598915B CN 202010422711 A CN202010422711 A CN 202010422711A CN 111598915 B CN111598915 B CN 111598915B
- Authority
- CN
- China
- Prior art keywords
- point
- points
- trunk
- seed
- horizontal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
-
- 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
Abstract
The application relates to a point cloud single wood segmentation method, a point cloud single wood segmentation device, point cloud single wood segmentation equipment and a computer readable medium. The method comprises the following steps: horizontally clustering the point cloud to be processed to obtain a plurality of horizontal clusters, wherein the point cloud to be processed is point cloud data obtained by scanning target trees by scanning equipment; extracting trunk seed points of a target tree from a plurality of horizontal clusters; constructing a path diagram in the point cloud to be processed according to trunk seed points; traversing the shortest path point from the trunk seed point in the path diagram; and merging the current shortest path point into a single wood point set where the trunk seed point is located under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is smaller than or equal to a distance threshold, wherein the single wood point set represents a set of points on single wood corresponding to the trunk seed point. According to the technical scheme, the influence of the cross of the branches among the trees on the single-tree segmentation is effectively reduced, the segmentation precision is improved, and the segmentation efficiency is accelerated.
Description
Technical Field
The application relates to the technical field of mapping point cloud data processing, in particular to a point cloud single wood segmentation method, a device, equipment and a computer readable medium.
Background
In reverse engineering, the method of obtaining point data sets on the appearance surface of an object by a measuring instrument has been widely applied to detection of point data in various fields of agriculture, forestry, ground disaster, electric power, mapping and the like. In particular, the point cloud data obtained by scanning the tree by using scanning equipment such as ground laser radar, mobile laser radar (LiDAR) and the like can directly obtain the tree height, the breast diameter, the crown area and the crown volume, and the biomass and the carbon reserve can be estimated by using an empirical formula based on a single tree structure.
Currently, there are many drawbacks in the related art when the point cloud data is subjected to single-tree segmentation. The inventor finds that the existing foundation laser radar single-wood segmentation method has the problems of low segmentation precision, low efficiency and the like, mainly has low tree identification accuracy, is easy to identify non-trees as trees according to simple slicing clustering so as to grow branches and leaves as seed points, is easy to be influenced by peripheral tree branches and leaves when the branches and leaves are subjected to cloud growth after the trunk identification, and has serious over-segmentation and missed segmentation, and single-wood parameter extraction precision and display effect are influenced. In the related art, a simple shortest path algorithm is adopted for growth, however, the inventor finds that the simple shortest path growth has the problems of low efficiency and the like in the research process, can not meet the requirements of production and the like, and can not process laser point clouds with more dense quantity.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The application provides a point cloud single wood segmentation method, a device, equipment and a computer readable medium, so as to solve the technical problems of low segmentation precision and low efficiency.
In a first aspect, the present application provides a method for splitting a point cloud mono tree, including: horizontally clustering the point cloud to be processed to obtain a plurality of horizontal clusters, wherein the point cloud to be processed is point cloud data obtained by scanning target trees by scanning equipment; extracting trunk seed points of a target tree from a plurality of horizontal clusters; constructing a path diagram in the point cloud to be processed according to trunk seed points; traversing the shortest path point from the trunk seed point in the path diagram; and merging the current shortest path point into a single wood point set where the trunk seed point is located under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is smaller than or equal to a distance threshold, wherein the single wood point set represents a set of points on single wood corresponding to the trunk seed point.
Optionally, before horizontally clustering the point cloud to be processed to obtain a plurality of horizontal clusters, the method further includes preprocessing the point cloud to be processed according to the following mode: denoising the point cloud to be processed to obtain a denoising point cloud; classifying the noise reduction point cloud to obtain ground points and tree points; the relative height of the ground point from the tree point is taken as the canopy height of the tree point.
Optionally, performing horizontal clustering on the point cloud to be processed to obtain a plurality of horizontal clusters includes: performing horizontal cutting on the tree points with the canopy height according to a preset height threshold value to obtain tree points with multiple horizontal layers; and carrying out horizontal European clustering on the tree points of each horizontal layer to obtain a plurality of horizontal clusters.
Optionally, extracting the trunk seed points of the target tree from the plurality of horizontal clusters comprises: calculating the gravity center points of each horizontal cluster, and adding non-trunk seed point marks for each gravity center point; taking the gravity center point in a layer of horizontal layer closest to the ground point as a first alternative trunk seed point, and adding a trunk seed point mark for the first alternative trunk seed point; calculating first directions of N gravity center points and first alternative trunk seed points in a target horizontal layer, wherein the target horizontal layer is other horizontal layers except the horizontal layer where the first alternative trunk seed points are located; calculating a second direction between any two gravity center points in N gravity center points of the target horizontal layer; under the condition that the included angle between the first direction and the ground is larger than an angle threshold value, taking a group of gravity points with the smallest included angle formed by the first direction and the second direction as second alternative trunk seed points, and adding trunk seed point marks for the second alternative trunk seed points; and taking the second alternative trunk seed point as a final trunk seed point under the condition that the second alternative trunk seed point meets the preset condition.
Optionally, before traversing the shortest path points from the trunk seed point in the path graph, creating a set of barycentric points as follows: performing European clustering of three-dimensional space on tree points to obtain a plurality of space clusters; selecting a plurality of target clusters from the plurality of space clusters, wherein the target clusters are space clusters with the number of cluster points larger than a screening threshold value; and calculating a barycenter point set consisting of barycenter points of the plurality of target clusters, wherein the plurality of barycenter points in the barycenter point set are ordered from large to small according to the number of cluster points in the target clusters where the barycenter points are located.
Optionally, in the case that the distance from the current shortest path point queried in the path diagram to the reference shortest path point adjacent to the current shortest path point is greater than the distance threshold, the method further includes re-querying the shortest path point as follows: inquiring the nearest gravity center point closest to the trunk seed point in the gravity center point set; and inquiring the shortest path point closest to the trunk seed point in the target cluster corresponding to the closest gravity center point, and merging the shortest path point into the single wood point set where the trunk seed point is located.
Optionally, after merging the current shortest path point into the single wood point set where the trunk seed point is located, the method further comprises: carrying out voxelized downsampling on the point cloud to be processed to obtain a voxel gravity center point; and constructing a k-dtree for the voxel gravity center point according to the single-wood segmentation relation represented by the single-wood point set so as to restore the single-wood mark to the point cloud to be processed in a nearest neighbor mode.
In a second aspect, the present application provides a point cloud single wood splitting device, including: the clustering module is used for horizontally clustering the point cloud to be processed to obtain a plurality of horizontal clusters, wherein the point cloud to be processed is point cloud data obtained by scanning the target tree by the scanning equipment; the extraction module is used for extracting trunk seed points of the target trees from the plurality of horizontal clusters; the construction module is used for constructing a path diagram in the point cloud to be processed according to the trunk seed points; the query module is used for traversing the shortest path point from the trunk seed point to be queried in the path diagram; and the distribution module is used for merging the current shortest path point into a single wood point set where the trunk seed point is located under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is smaller than or equal to a distance threshold value, wherein the single wood point set represents a set of points on single wood corresponding to the trunk seed point.
In a third aspect, the present application provides a computer device comprising a memory, a processor, the memory having stored thereon a computer program executable on the processor, the processor executing the steps of the method of any of the first aspects described above.
In a fourth aspect, the present application also provides a computer readable medium, the program code causing a processor to perform any of the methods of the first aspect.
Compared with the related art, the technical scheme provided by the embodiment of the application has the following advantages:
the method comprises the steps of horizontally clustering point clouds to be processed to obtain a plurality of horizontal clusters, wherein the point clouds to be processed are point cloud data obtained by scanning target trees by scanning equipment; extracting trunk seed points of a target tree from a plurality of horizontal clusters; constructing a path diagram in the point cloud to be processed according to trunk seed points; traversing the shortest path point from the trunk seed point in the path diagram; under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is smaller than or equal to a distance threshold value, merging the current shortest path point into a single-tree point set where a trunk seed point is located, wherein the single-tree point set represents a point cloud single-tree segmentation method belonging to a set of points on single tree corresponding to the trunk seed point, the influence of inter-branch crossing of a plurality of trees on single-tree segmentation is effectively reduced, segmentation precision is improved, and segmentation efficiency is accelerated.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described, and it is obvious to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort.
Fig. 1 is a schematic diagram of an optional hardware environment of a point cloud single-tree segmentation method according to an embodiment of the present application;
fig. 2 is a flowchart of an alternative method for dividing a point cloud single wood according to an embodiment of the present application;
FIG. 3 is an optional flow chart of preprocessing point cloud data according to an embodiment of the present application;
fig. 4 is an alternative trunk seed point extraction flow chart provided according to an embodiment of the present application;
fig. 5 is a block diagram of an alternative point cloud single-tree splitting device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present application based on the embodiments herein.
In the following description, suffixes such as "module", "component", or "unit" for representing elements are used only for facilitating the description of the present application, and are not of specific significance per se. Thus, "module" and "component" may be used in combination.
According to an aspect of the embodiments of the present application, an embodiment of a point cloud single-tree segmentation method is provided.
Alternatively, in the embodiment of the present application, the above-described point cloud mono segmentation method may be applied to a hardware environment configured by the terminal 101 and the server 103 as shown in fig. 1. As shown in fig. 1, the server 103 is connected to the terminal 101 through a network, which may be used to provide services to the terminal or a client installed on the terminal, and a database 105 may be provided on the server or independent of the server, for providing data storage services to the server 103, where the network includes, but is not limited to: a wide area network, metropolitan area network, or local area network, and terminal 101 includes, but is not limited to, a PC, a cell phone, a tablet computer, etc.
A point cloud single tree segmentation method in the embodiment of the present application (i.e., the point cloud segmentation method in the present application is actually a fast point cloud single tree segmentation method based on a shortest path and an european clustering algorithm) may be executed by the server 103, as shown in fig. 2, and the method may include the following steps:
step S201, horizontally clustering the point cloud to be processed to obtain a plurality of horizontal clusters;
in this embodiment of the present application, the point cloud to be processed is point cloud data obtained by scanning a target tree by using a scanning device, where the scanning device may be a ground laser radar, a backpack or a vehicle-mounted laser radar. When the huge point cloud data is segmented into single woods, trunk seed points are required to be extracted firstly to represent the main body of the single woods, the point cloud data are often gathered, horizontal clustering can be carried out, and a plurality of horizontal clusters are obtained so as to find the trunk seed points. The method can also perform voxel downsampling on the point cloud to be processed, and the point cloud is represented by the voxel gravity center point, so that the data processing efficiency can be effectively improved.
Step S202, extracting trunk seed points of a target tree from a plurality of horizontal clusters;
in the embodiment of the application, a representative point can be selected from the horizontal clusters obtained by clustering to serve as a trunk seed point of the single tree, and the representative point can be represented by calculating the gravity center point of the horizontal clusters.
Step S203, constructing a path diagram in the point cloud to be processed according to the trunk seed points;
in this embodiment, after the trunk of the tree, that is, the trunk seed point is obtained, the trunk seed point may be used as an alternative point, and a path diagram for dividing the trunk seed point to which each point cloud belongs is constructed in the point cloud to be processed, that is, the original point cloud, so as to perform shortest path growth on the tree.
Step S204, traversing the shortest path point from the trunk seed point in the path diagram;
step S205, merging the current shortest path point into the single wood point set where the trunk seed point is located under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is less than or equal to the distance threshold value.
In the embodiment of the application, the single-tree point set represents a set of points on a single tree corresponding to trunk seed points, namely, points belonging to the same tree are classified into one single-tree point set. The distance threshold may be the maximum distance allowed by two adjacent points in the point cloud that make up the same tree. The point is closest to a point on which tree of all trees is considered to be the same tree as the tree, and the same ID is marked, namely the point is divided into a single-tree point set where the trunk seed point closest to the point is located. The points in each tree will mark the total path and path distances from other points.
In the embodiment of the application, after the current shortest path point is classified into a single wood point set, the point is inserted into a trunk seed point and a path diagram is updated to serve as a reference shortest path point when the next shortest path point is inquired.
Optionally, the embodiment of the present application provides a method for preprocessing point cloud data before horizontally clustering point clouds to be processed to obtain a plurality of horizontal clusters, as shown in fig. 3, including the following steps:
step S301, denoising the point cloud to be processed to obtain a noise reduction point cloud;
step S302, classifying the noise reduction point cloud to obtain ground points and tree points;
in step S303, the relative height of the ground point from the tree point is taken as the canopy height of the tree point.
In this embodiment of the present application, the point cloud to be processed includes all objects in the scanning area, even includes noise point data, and after the original point cloud data is obtained, the point cloud data needs to be denoised and classified, which may be classified into ground points and tree points, where the ground points represent that actual objects corresponding to the point cloud data are ground, the tree points represent that actual objects corresponding to the point cloud data are trees, and after classification, the point cloud data only change category attributes. The relative height of the ground point from the tree point is used as the canopy height of the tree point, namely, the classified point cloud data is normalized, namely, the Z value of the tree point is subtracted from the Z value of the ground point to obtain the canopy height value of the tree point, so that the influence of topography fluctuation on the subsequent single-tree segmentation is reduced.
The normalization may be performed by using ground point normalization or by using elevation digital model (DEM).
Optionally, performing horizontal clustering on the point cloud to be processed to obtain a plurality of horizontal clusters may include the following steps:
step 1, horizontally slicing the tree points at the height of the canopy according to a preset height threshold to obtain tree points of a plurality of horizontal layers;
and 2, performing horizontal European clustering on the tree points of each horizontal layer to obtain a plurality of horizontal clusters.
In this embodiment, only consider the canopy height of trees point when carrying out the horizontal cutting layer to the trees point to reduce the influence that topography fluctuation brought, the more the number of cutting layer is, the more can reduce the error rate of discernment trees promptly, the more the number of cutting layer is, and other short shrubs or other ground object's influence is less. As a preferred scheme of the embodiment of the application, the number of the cutting layers is not less than 5 layers, and the cutting threshold height is not less than 3 times of the average point spacing.
In this embodiment of the present application, the tree points of each horizontal layer are horizontally clustered, specifically, the tree points of each horizontal layer may be horizontally European clustered, that is, european clustered according to two-dimensional XY directions, to obtain a plurality of horizontal clusters.
Optionally, the embodiment of the present application provides an optional extraction method of trunk seed points, as shown in fig. 4, step S202 may include:
step S401, calculating the gravity center points of each horizontal cluster, and adding non-trunk seed point marks for each gravity center point;
step S402, taking a gravity center point in a layer of horizontal layers closest to the ground point as a first alternative trunk seed point, and adding a trunk seed point mark for the first alternative trunk seed point;
step S403, calculating first directions of N gravity center points and first alternative trunk seed points in a target horizontal layer, wherein the target horizontal layer is other horizontal layers except the horizontal layer where the first alternative trunk seed points are located;
step S404, calculating a second direction between any two gravity center points in N gravity center points of the target horizontal layer;
step S405, under the condition that the included angle between the first direction and the ground is larger than an angle threshold value, taking a group of gravity points with the smallest included angle formed by the first direction and the second direction as second alternative trunk seed points, and adding trunk seed point marks for the second alternative trunk seed points;
in step S406, if the second alternative trunk seed point meets the preset condition, the second alternative trunk seed point is taken as the final trunk seed point.
In the embodiment of the application, when trunk seed points of a target tree are extracted from a plurality of horizontal clusters, first, calculating the gravity center points of each horizontal cluster in each horizontal layer, and adding all the gravity center points at the moment with non-trunk seed point marks. A barycentric point with a non-trunk seed point identification indicates that the barycentric point is not a trunk seed point. The center of gravity point may be calculated by summing each of the three-dimensional component X, Y and Z values in a horizontal cluster separately divided by the total number of points.
In the embodiment of the application, a layer of horizontal layer closest to the ground is taken as a layer 0, point cloud data in the layer is taken as a layer 0 data, and the like. And taking the gravity center point of each horizontal cluster in the 0-layer data as a first alternative trunk seed point which is an alternative trunk point, and changing the non-trunk seed point mark added in the step before the gravity center point into a trunk seed point mark so as to indicate that the gravity center point can be used as a trunk seed point. Because the closer the tree is to the ground, the more likely it is the trunk, according to common sense of life.
In this embodiment, N points are searched up layer by layer from a layer 0 to each gravity center point (i.e., a first alternative trunk seed point of the layer 0), a first direction a between the gravity center point and the searched N points and a second direction Bi of any two points in the N points are calculated, an included angle between the first direction a and the second direction Bi is minimum, a group of trunk points with an included angle between the first direction a and the ground being greater than a certain angle threshold is used as a second alternative trunk point, an original non-trunk seed point identifier is changed into a trunk seed point identifier, and each point in the N points is calculated in sequence. According to the tree growth mode, the trunk has direction consistency, so that the probability that the tree is the same tree and is the trunk is maximum when the included angle between the first direction A and the second direction Bi is minimum, meanwhile, the tree growth can form a certain included angle with the ground, the vertical growth is a 90-degree included angle under normal conditions, and the threshold of the angle, namely the threshold of the included angle between the tree and the ground, can be set to be 30 degrees, so that errors are reduced when the tree is identified.
In this embodiment of the present application, after the preliminary screening is completed, screening is performed on all the second alternative trunk seed points, which may be to reserve, as the final trunk seed point, the center of gravity point that is greater than K points and has a height that is greater than a certain threshold.
Optionally, before traversing the shortest path points from the trunk seed point in the path graph, creating a set of barycentric points as follows:
step 1, performing European clustering of three-dimensional space on tree points to obtain a plurality of space clusters;
step 2, selecting a plurality of target clusters from a plurality of space clusters, wherein the target clusters are space clusters with the number of cluster points larger than a screening threshold value;
and 3, calculating a barycenter point set consisting of barycenter points of the plurality of target clusters, wherein the plurality of barycenter points in the barycenter point set are ordered from large to small according to the number of cluster points in the target clusters where the barycenter points are located.
In the embodiment of the application, the point cloud data can be spatially three-dimensional European clustering to accelerate data processing, specifically, the tree points are spatially European clustered to obtain a plurality of spatial clusters, and the spatial clusters with the clustering points larger than the screening threshold are screened out, so that shortest path calculation is not performed on the spatial clusters with fewer clustering points in the subsequent steps. Alternatively, the screening threshold may be 500 points.
In the embodiment of the application, after the spatial clusters with the cluster points larger than the screening threshold value are screened out, the gravity center points of the spatial clusters are required to be calculated, and the gravity center points are formed into a gravity center point set, wherein the sorting mode of the gravity center points in the gravity center point set sorts the cluster points from large to small according to the number of the cluster points in the spatial clusters where the gravity center points are located, so that most of the spatial clusters are preferentially solved when the shortest path points are inquired, and the efficiency is higher.
Optionally, in the case that the distance from the current shortest path point queried in the path diagram to the reference shortest path point adjacent to the current shortest path point is greater than the distance threshold, the method further includes re-querying the shortest path point as follows:
step 1, inquiring the nearest gravity center point closest to a trunk seed point in the gravity center point set;
and 2, inquiring the shortest path point closest to the trunk seed point in the target cluster corresponding to the closest gravity point, and merging the shortest path point into the single wood point set where the trunk seed point is located.
In the embodiment of the application, when the distance from the queried current shortest path point to the adjacent reference shortest path point is larger than the maximum distance allowed by two adjacent points in the point cloud forming the same tree, the nearest gravity center point closest to the trunk seed point is queried from the gravity center point set, and the shortest path point closest to the trunk seed point is queried again in the space cluster where the nearest gravity center point is located.
Optionally, after merging the current shortest path point into the single wood point set where the trunk seed point is located, the method further comprises:
step 1, carrying out voxelized downsampling on a point cloud to be processed to obtain a voxel gravity center point;
and 2, constructing a k-d tree for the voxel gravity center point according to a single-tree segmentation relation represented by the single-tree point set so as to restore the single-tree mark to the point cloud to be processed in a nearest neighbor mode.
In the embodiment of the application, after tree growth is performed on all the down-sampled points, the down-sampled points need to be restored to the original point cloud, and the Yun Shan tree marks of the original points can be restored by using a nearest neighbor query method.
The method comprises the steps of horizontally clustering point clouds to be processed to obtain a plurality of horizontal clusters, wherein the point clouds to be processed are point cloud data obtained by scanning target trees by scanning equipment; extracting trunk seed points of a target tree from a plurality of horizontal clusters; constructing a path diagram in the point cloud to be processed according to trunk seed points; traversing the shortest path point from the trunk seed point in the path diagram; under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is smaller than or equal to a distance threshold value, merging the current shortest path point into a single-tree point set where a trunk seed point is located, wherein the single-tree point set represents a point cloud single-tree segmentation method belonging to a set of points on single tree corresponding to the trunk seed point, the influence of inter-branch crossing of a plurality of trees on single-tree segmentation is effectively reduced, segmentation precision is improved, and segmentation efficiency is accelerated.
According to still another aspect of the embodiments of the present application, as shown in fig. 5, there is provided a point cloud mono segmentation apparatus, including: the clustering module 501 is configured to perform horizontal clustering on a point cloud to be processed to obtain a plurality of horizontal clusters, where the point cloud to be processed is point cloud data obtained by scanning a target tree by a scanning device; an extraction module 502, configured to extract trunk seed points of a target tree from a plurality of horizontal clusters; a construction module 503, configured to construct a path diagram in the point cloud to be processed according to the trunk seed points; a query module 504, configured to traverse a shortest path point from the trunk seed point in the path graph; the allocation module 505 is configured to merge the current shortest path point into a set of single-tree points where the trunk seed point is located, where the set of single-tree points represents a set of points on a single tree corresponding to the trunk seed point, if a distance from the current shortest path point to a reference shortest path point adjacent to the current shortest path point is less than or equal to a distance threshold.
It should be noted that, the clustering module 501 in this embodiment may be used to perform step S201 in this embodiment, the extracting module 502 in this embodiment may be used to perform step S202 in this embodiment, the building module 503 in this embodiment may be used to perform step S203 in this embodiment, the query module 504 in this embodiment may be used to perform step S204 in this embodiment, and the allocating module 505 in this embodiment may be used to perform step S205 in this embodiment.
The point cloud single wood segmentation method (namely the rapid point cloud single wood segmentation method based on the shortest path and the European clustering algorithm) provided by the embodiment of the invention is used for carrying out ground point classification on the obtained point cloud to be segmented and then normalizing; voxel downsampling is carried out on point cloud data, and the center of gravity point of the voxels is used for replacing the original point cloud; extracting data from the ground to a preset height range from the normalized data, and extracting trunk point clouds as tree growth seed points; european clustering is carried out on the data according to a set threshold value, and a path diagram is constructed on the point cloud data; and taking the trunk seed point cloud as an alternative point, traversing the shortest path point of the query distance seed point, and traversing the alternative point in a circulating way until all points are merged into the target point set.
It should be noted that the above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments. It should be noted that the above modules may be implemented in software or hardware as a part of the apparatus in the hardware environment shown in fig. 1.
Optionally, the point cloud single wood splitting device further includes: the denoising module is used for denoising the point cloud to be processed to obtain a denoising point cloud; the classification module is used for classifying the noise reduction point cloud to obtain ground points and tree points; and the normalization module is used for taking the relative height of the ground point from the tree point as the canopy height of the tree point.
Optionally, the point cloud single wood splitting device further includes: the horizontal layer cutting module is used for horizontally cutting the tree points at the height of the canopy according to a preset height threshold value to obtain tree points of a plurality of horizontal layers; and the horizontal European clustering module is used for carrying out horizontal European clustering on the tree points of each horizontal layer to obtain a plurality of horizontal clusters.
Optionally, the point cloud single wood splitting device further includes: the gravity center point calculating module is used for calculating the gravity center points of each horizontal cluster and adding non-trunk seed point marks for each gravity center point; the first alternative trunk seed point selection module is used for taking the gravity center point in a layer of horizontal layer closest to the ground point as a first alternative trunk seed point and adding a trunk seed point mark for the first alternative trunk seed point; the first direction calculation module is used for calculating first directions of N center of gravity points and first alternative trunk seed points in a target horizontal layer, wherein the target horizontal layer is other horizontal layers except for the horizontal layer where the first alternative trunk seed points are located; the second direction calculation module is used for calculating a second direction between any two gravity center points in the N gravity center points of the target horizontal layer; the second alternative trunk seed point selection module is used for taking a group of gravity points with the smallest included angle formed by the first direction and the second direction as second alternative trunk seed points and adding trunk seed point marks for the second alternative trunk seed points under the condition that the included angle between the first direction and the ground is larger than an angle threshold value; and the confirmation module is used for taking the second alternative trunk seed point as a final trunk seed point under the condition that the second alternative trunk seed point meets the preset condition.
Optionally, the point cloud single wood splitting device further includes: the spatial European clustering module is used for carrying out European clustering on tree points in a three-dimensional space to obtain a plurality of spatial clusters; the screening module is used for selecting a plurality of target clusters from the plurality of space clusters, wherein the target clusters are space clusters with the number of cluster points larger than a screening threshold value; the center of gravity point set calculation module is used for calculating a center of gravity point set formed by center of gravity points of a plurality of target clusters, wherein the plurality of center of gravity points in the center of gravity point set are ordered from large to small according to the number of cluster points in the target clusters where the center of gravity points are located.
Optionally, the point cloud single wood splitting device further includes: the nearest gravity center point searching module is used for searching the nearest gravity center point closest to the trunk seed point in the gravity center point set; and the shortest path point searching module is used for searching the shortest path point closest to the trunk seed point in the target cluster corresponding to the closest gravity point, and merging the shortest path point into the single wood point set where the trunk seed point is located.
Optionally, the point cloud single wood splitting device further includes: the voxelized downsampling module is used for carrying out voxelized downsampling on the point cloud to be processed to obtain a voxel gravity center point; and the proximity query module is used for constructing a k-dtree for the voxel gravity center point according to the single-wood segmentation relation represented by the single-wood point set so as to restore the single-wood mark for the point cloud to be processed in a nearest-neighbor mode.
The point cloud single wood segmentation method (namely the rapid point cloud single wood segmentation method based on the shortest path and the European clustering algorithm) provided by the embodiment of the invention also has the following technical advantages, for example: 1. the method has the advantages that the method is particularly effective in dense forest areas, compared with other algorithms, the effect is more obvious, 2, the single-tree segmentation efficiency is obviously improved through early voxel resampling and European clustering (namely, the shortest path can be effectively solved, but the space complexity and the time complexity of a point-by-point calculation mode are higher under the principle of the shortest path algorithm, voxel formation can effectively reduce the over-dense area without influencing the overall distribution, and European clustering can effectively improve the efficiency of the shortest path growth process).
There is also provided in accordance with a further aspect of embodiments of the present application a computer device comprising a memory, a processor, the memory storing a computer program executable on the processor, the processor implementing the above steps when executing the computer program.
The memory and the processor in the computer device communicate with the communication interface through a communication bus. The communication bus may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The communication bus may be classified as an address bus, a data bus, a control bus, or the like.
The memory may include random access memory (Random Access Memory, RAM) or non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.
The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processing, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
There is also provided, in accordance with yet another aspect of an embodiment of the present application, a computer-readable medium.
Optionally, in an embodiment of the present application, the computer readable medium is configured to store program code for the processor to perform the steps of:
step S201, horizontally clustering the point cloud to be processed to obtain a plurality of horizontal clusters;
step S202, extracting trunk seed points of a target tree from a plurality of horizontal clusters;
step S203, constructing a path diagram in the point cloud to be processed according to the trunk seed points;
step S204, traversing the shortest path point from the trunk seed point in the path diagram;
step S205, merging the current shortest path point into the single wood point set where the trunk seed point is located under the condition that the distance from the current shortest path point to the adjacent reference shortest path point is less than or equal to the distance threshold value.
Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.
In specific implementation, the embodiments of the present application may refer to the above embodiments, which have corresponding technical effects.
It is to be understood that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For a hardware implementation, the processing units may be implemented within one or more application specific integrated circuits (Application Specific Integrated Circuits, ASIC), digital signal processors (Digital Signal Processing, DSP), digital signal processing devices (DSP devices, DSPD), programmable logic devices (Programmable Logic Device, PLD), field programmable gate arrays (Field-Programmable Gate Array, FPGA), general purpose processors, controllers, microcontrollers, microprocessors, other electronic units configured to perform the functions described herein, or a combination thereof.
For a software implementation, the techniques described herein may be implemented by means of units that perform the functions described herein. The software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially or, what contributes to the prior art, or part of the technical solutions, may be embodied in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc. It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is merely a specific embodiment of the application to enable one skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (4)
1. The method for dividing the point cloud single wood is characterized by comprising the following steps of:
performing horizontal clustering on point clouds to be processed to obtain a plurality of horizontal clusters, wherein the point clouds to be processed are point cloud data obtained by scanning target trees by scanning equipment;
extracting trunk seed points of the target tree from a plurality of the horizontal clusters;
constructing a path diagram in the point cloud to be processed according to the trunk seed points;
traversing shortest path points from the trunk seed points in the path diagram;
merging the current shortest path point into a single wood point set where the trunk seed point is located under the condition that the distance from the current shortest path point to a reference shortest path point adjacent to the current shortest path point is smaller than or equal to a distance threshold, wherein the single wood point set represents a set of points on single wood corresponding to the trunk seed point;
carrying out horizontal clustering on the point cloud to be processed, and preprocessing the point cloud to be processed according to the following mode before obtaining a plurality of horizontal clusters:
denoising the point cloud to be processed to obtain a denoising point cloud;
classifying the noise reduction point cloud to obtain ground points and tree points;
taking the relative height of the ground point from the tree point as the canopy height of the tree point;
performing horizontal clustering on the point cloud to be processed to obtain a plurality of horizontal clusters comprises:
performing horizontal cutting on the tree points at the height of the canopy according to a preset height threshold to obtain the tree points of a plurality of horizontal layers;
performing horizontal European clustering on the tree points of each horizontal layer to obtain a plurality of horizontal clusters;
extracting trunk seed points of the target tree from the plurality of horizontal clusters includes:
calculating the gravity center points of the horizontal clusters, and adding non-trunk seed point marks for the gravity center points;
taking the gravity center point in a layer of horizontal layers closest to the ground point as a first alternative trunk seed point, and adding a trunk seed point mark for the first alternative trunk seed point;
calculating first directions of N gravity center points and the first alternative trunk seed points in a target horizontal layer, wherein the target horizontal layer is other horizontal layers except the horizontal layer where the first alternative trunk seed points are located;
calculating a second direction between any two of the gravity center points of the N target horizontal layers;
when the included angle between the first direction and the ground is larger than an angle threshold value, taking a group of gravity points with the smallest included angle formed by the first direction and the second direction as second alternative trunk seed points, and adding the trunk seed point identification for the second alternative trunk seed points;
taking the second alternative trunk seed point as the final trunk seed point under the condition that the second alternative trunk seed point meets the preset condition;
before traversing the shortest path points from the trunk seed points in the path graph, creating a center of gravity point set according to the following mode:
performing European clustering of the three-dimensional space on the tree points to obtain a plurality of space clusters;
selecting a plurality of target clusters from the plurality of spatial clusters, wherein the target clusters are spatial clusters with the number of cluster points larger than a screening threshold value;
calculating the gravity center point set formed by gravity center points of the plurality of target clusters, wherein the plurality of gravity center points in the gravity center point set are ordered from large to small according to the number of cluster points in the target clusters where the gravity center points are located.
2. The method according to claim 1, wherein in case the distance from the current shortest path point queried in the path graph to the reference shortest path point adjacent thereto is greater than the distance threshold, further comprising re-querying the shortest path point as follows:
searching the nearest gravity center point closest to the trunk seed point in the gravity center point set;
inquiring the shortest path point closest to the trunk seed point in the target cluster corresponding to the closest center of gravity point, and merging the shortest path point into a single wood point set where the trunk seed point is located.
3. The method of claim 2, wherein after merging the current shortest path point into the set of single wood points in which the trunk seed point is located, the method further comprises:
carrying out voxelized downsampling on the point cloud to be processed to obtain a voxel gravity center point;
and constructing a k-dtree for the voxel gravity center point according to the single-wood segmentation relation represented by the single-wood point set so as to restore the single-wood mark for the point cloud to be processed in a nearest neighbor mode.
4. A computer device comprising a memory, a processor, the memory having stored therein a computer program executable on the processor, characterized in that the processor, when executing the computer program, implements the steps of the method of any of the preceding claims 1 to 3.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010422711.3A CN111598915B (en) | 2020-05-19 | 2020-05-19 | Point cloud single wood segmentation method, device, equipment and computer readable medium |
| PCT/CN2020/092868 WO2021232467A1 (en) | 2020-05-19 | 2020-05-28 | Point cloud single-tree segmentation method and apparatus, device and computer-readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010422711.3A CN111598915B (en) | 2020-05-19 | 2020-05-19 | Point cloud single wood segmentation method, device, equipment and computer readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111598915A CN111598915A (en) | 2020-08-28 |
| CN111598915B true CN111598915B (en) | 2023-06-30 |
Family
ID=72192439
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010422711.3A Active CN111598915B (en) | 2020-05-19 | 2020-05-19 | Point cloud single wood segmentation method, device, equipment and computer readable medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111598915B (en) |
| WO (1) | WO2021232467A1 (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112308839B (en) * | 2020-10-31 | 2022-08-02 | 云南师范大学 | Single-tree segmentation method and device for natural forest, computer equipment and storage medium |
| CN113095571B (en) * | 2021-04-14 | 2022-08-09 | 河北农业大学 | Biomass model prediction method and system based on tree metabolic rate theory |
| CN114387288A (en) * | 2022-01-17 | 2022-04-22 | 滁州学院 | Single standing tree three-dimensional information extraction method based on vehicle-mounted laser radar point cloud data |
| CN114510771B (en) * | 2022-04-07 | 2022-07-08 | 深圳小库科技有限公司 | Method, device and equipment for generating mold of building curved surface and computer readable medium |
| CN114677381B (en) * | 2022-05-30 | 2022-10-11 | 天津云圣智能科技有限责任公司 | Insulator segmentation method and device, electronic equipment and computer readable storage medium |
| CN115512244B (en) * | 2022-08-19 | 2024-01-05 | 中国林业科学研究院资源信息研究所 | Method and system for determining carbon reserves of single tree |
| CN115953607B (en) * | 2023-01-04 | 2024-02-13 | 北京数字绿土科技股份有限公司 | Trunk breast diameter extraction method and system using point cloud data |
| CN116310849B (en) * | 2023-05-22 | 2023-09-19 | 深圳大学 | Tree point cloud monomerization extraction method based on three-dimensional morphological characteristics |
| CN117252863B (en) * | 2023-11-13 | 2024-02-09 | 山东省地质测绘院 | Quick detection and analysis method for geographic information abnormal data |
| CN117437269B (en) * | 2023-12-22 | 2024-04-16 | 深圳大学 | Tree point cloud non-rigid registration method and related equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101922914A (en) * | 2010-08-27 | 2010-12-22 | 中国林业科学研究院资源信息研究所 | Crown information extraction method and system based on high spatial resolution remote sense image |
| CN105894044A (en) * | 2016-04-29 | 2016-08-24 | 武汉大学 | Single-plant tree point cloud automatic extraction method based on vehicle-mounted laser scanning data |
| CN109712229A (en) * | 2018-11-26 | 2019-05-03 | 漳州通正勘测设计院有限公司 | A kind of isolated tree wooden frame point extracting method, device, equipment and storage medium |
| CN110992473A (en) * | 2019-12-13 | 2020-04-10 | 武汉大学 | Tree branch modeling method and system based on vehicle-mounted laser scanning point cloud |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2528669B (en) * | 2014-07-25 | 2017-05-24 | Toshiba Res Europe Ltd | Image Analysis Method |
-
2020
- 2020-05-19 CN CN202010422711.3A patent/CN111598915B/en active Active
- 2020-05-28 WO PCT/CN2020/092868 patent/WO2021232467A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101922914A (en) * | 2010-08-27 | 2010-12-22 | 中国林业科学研究院资源信息研究所 | Crown information extraction method and system based on high spatial resolution remote sense image |
| CN105894044A (en) * | 2016-04-29 | 2016-08-24 | 武汉大学 | Single-plant tree point cloud automatic extraction method based on vehicle-mounted laser scanning data |
| CN109712229A (en) * | 2018-11-26 | 2019-05-03 | 漳州通正勘测设计院有限公司 | A kind of isolated tree wooden frame point extracting method, device, equipment and storage medium |
| CN110992473A (en) * | 2019-12-13 | 2020-04-10 | 武汉大学 | Tree branch modeling method and system based on vehicle-mounted laser scanning point cloud |
Non-Patent Citations (2)
| Title |
|---|
| 机载激光雷达森林资源调查系统的设计与试验;瞿帅 等;《西北林学院学报》;20180831;第1.1，1.3.2.1章节 * |
| 风力载荷下树木安全性评估;黄笑;《中国优秀硕士学位论文全文数据库 农业科技辑》;20200515(第05期);第4.2章节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021232467A1 (en) | 2021-11-25 |
| CN111598915A (en) | 2020-08-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111598915B (en) | Point cloud single wood segmentation method, device, equipment and computer readable medium | |
| CN110363861B (en) | Laser radar point cloud-based field crop three-dimensional reconstruction method | |
| CN110334391B (en) | Automatic planning method for collecting circuit of multi-dimensional constraint wind power plant | |
| CN111462134A (en) | Single-tree segmentation method and system for fusing high-resolution remote sensing image and laser radar point cloud | |
| CN113538264B (en) | Denoising method and device for point cloud data and storage medium | |
| CN110717496B (en) | Complex scene tree detection method based on neural network | |
| CN116310849B (en) | Tree point cloud monomerization extraction method based on three-dimensional morphological characteristics | |
| CN104778281A (en) | Image index parallel construction method based on community analysis | |
| Miao et al. | Label3DMaize: toolkit for 3D point cloud data annotation of maize shoots | |
| CN111861933A (en) | Point cloud denoising method and device based on space division | |
| CN115390040A (en) | Tree point cloud branch and leaf separation method based on segmentation geometric features | |
| CN111860359B (en) | Point cloud classification method based on improved random forest algorithm | |
| CN110222742B (en) | Point cloud segmentation method, device, storage medium and equipment based on layered multi-echo | |
| CN116385801A (en) | Tree classification method based on multispectral deep learning and point cloud technology | |
| CN116721228A (en) | Building elevation extraction method and system based on low-density point cloud | |
| CN115311576A (en) | Natural village boundary extraction method based on rural house distribution | |
| CN113724343B (en) | Tree point cloud compression method based on structure perception | |
| CN116721113A (en) | Three-dimensional point cloud plane segmentation method and system | |
| CN114511572A (en) | Single tree segmentation flow method for forest tree measurement | |
| Pan et al. | Multi-agent evolutionary clustering algorithm based on manifold distance | |
| CN112669461A (en) | Airport clearance safety detection method and device, electronic equipment and storage medium | |
| CN111783648A (en) | Method and device for extracting guardrail in road point cloud | |
| CN115272733B (en) | Trunk information extraction method based on point cloud data and multi-feature fusion | |
| CN116844649B (en) | Interpretable cell data analysis method based on gene selection | |
| CN115018893B (en) | Automatic building detail structure unitization method and system and readable storage medium |
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 | ||
| CB02 | Change of applicant information |
Address after: Room 2301-2308, third floor, building 2, incubator, Zhongguancun Software Park, Dongbeiwang, Haidian District, Beijing 100094 Applicant after: Beijing Digital Green Earth Technology Co.,Ltd. Address before: Room 2301-2308, third floor, building 2, incubator, Zhongguancun Software Park, Dongbeiwang, Haidian District, Beijing 100094 Applicant before: BEIJING GREENVALLEY TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |