CN116129391A - Method and system for extracting pavement tree from vehicle-mounted laser point cloud - Google Patents
Method and system for extracting pavement tree from vehicle-mounted laser point cloud Download PDFInfo
- Publication number
- CN116129391A CN116129391A CN202310409469.XA CN202310409469A CN116129391A CN 116129391 A CN116129391 A CN 116129391A CN 202310409469 A CN202310409469 A CN 202310409469A CN 116129391 A CN116129391 A CN 116129391A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- points
- extracting
- crown
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention belongs to the technical field of surveying and mapping science, and provides a method and a system for extracting a street tree from a vehicle-mounted laser point cloud. The method comprises the steps of acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data; performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data; dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value; and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data. The invention realizes accurate and rapid identification of the pavement tree.
Description
Technical Field
The invention belongs to the technical field of surveying and mapping science, and particularly relates to a method and a system for extracting a street tree from a vehicle-mounted laser point cloud.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The traditional street tree measurement utilizes the total station, real-time dynamic carrier phase difference (RTK) and other traditional mapping technologies to combine manual field investigation and tape measurement modes to acquire points and information, and the measurement modes and management means are time-consuming and labor-consuming and can not meet the requirements of rapid data updating and informationized management.
At present, the method for extracting the street tree based on the vehicle-mounted laser point cloud generally has the problems of low universality and low calculation efficiency.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a method and a system for extracting a street tree from a vehicle-mounted laser point cloud, which are used for classifying and identifying the street tree point cloud by utilizing vehicle-mounted laser point cloud data and a morphological method, and meanwhile, the accurate and rapid identification of the street tree is realized based on the rapid statistics of the information such as the tree diameter, the crown amplitude and the like of the street tree point cloud.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a first aspect of the invention provides a method of extracting a street tree from a vehicle-mounted laser point cloud.
A method of extracting a street tree from an onboard laser point cloud, comprising:
acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data;
performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data;
dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value;
and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
Further, the filtering process of the vehicle-mounted laser point cloud data comprises the following steps:
taking points with the z value of the local laser scanning coordinate smaller than a set parameter in vehicle-mounted laser point cloud data to form a point set P;
setting any laser point in P as the point, taking two adjacent left and right points on the same scanning line with the point, the point closest to the point on the previous scanning line and the next scanning line of the point and the two adjacent left and right points respectively to form a 3X 3 lattice;
calculating the Z-value convolution sum sigma Z of the 3X 3 lattice;
if sigma Z is smaller than the set threshold Z sum Marking all points in the dot matrix;
traversing all laser points in P, skipping points which are marked and not processing any more, and repeating the processes of constructing a dot matrix, calculating a z-value convolution sum and marking if the points are not marked;
the marked points in the vehicle-mounted laser point cloud data are ground point cloud data, and the unmarked points are non-ground point cloud data.
Further, the process of meshing the non-ground point cloud data by adopting the projection weighing method comprises the following steps:
calculating the extreme value of the abscissa and the ordinate of the non-ground point cloud data;
constructing a grid according to the extreme values of the abscissa and the ordinate of the non-ground point cloud data, and storing the non-ground point cloud data into the grid;
and selecting a grid with the grid density larger than a preset grid density threshold by adopting a plane projection weighing method.
Further, the fitting is performed on the point cloud data in the mesh forming the cylinder, and the process of extracting the rod-shaped object point cloud data comprises the following steps:
randomly selecting a subset S from a data set S obtained after projection weighing, wherein S is an assumed interior point;
estimating model parameters according to the subset s;
traversing all data except subset S in data set S, marking as an interior point if the data point is within a given error e, otherwise marking as an exterior point;
if the number of the consistent concentrated points meets a given threshold T, re-estimating model parameters by using all the internal points in the consistent concentrated points, and ending the algorithm;
if the number of the inner points in the consistent set is less than the threshold value T, reselecting a new subset s, and reselecting the subset s, estimating model parameters, marking and judging the threshold value;
after K iterations, selecting a consistent set with the largest number of internal points, re-estimating model parameters by using all the internal points in the consistent set, and ending the algorithm;
and obtaining the identified cylinder through RANSAC fitting of the point cloud data.
Further, the process of extracting the tree crown point cloud data comprises the following steps:
dividing three-dimensional grids for non-ground point cloud data;
traversing grids one by one, and marking as 1 if laser points exist in the current grids;
counting the number NG of grids marked as 1 in the range of 3 multiplied by 3 around each grid;
if NG > threshold TN, marking all points in the grid as tree crown points;
and classifying crown points on the specific trunk according to the trunk position to form an independent tree.
Further, the process of obtaining the street tree single point cloud data comprises the following steps:
clustering the crown point cloud data and the rod-shaped object point cloud data by adopting a DBSCAN algorithm, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
Further, the method further comprises: crown values are calculated according to the crown point cloud data, and tree diameter values are calculated according to the shaft point cloud data.
A second aspect of the invention provides a system for extracting a street tree from a vehicle-mounted laser point cloud.
A system for extracting a street tree from an onboard laser point cloud, comprising:
a data acquisition and separation module configured to: acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data;
a shaft extraction module configured to: performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data;
a crown extraction module configured to: dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value;
an output module configured to: and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
A third aspect of the present invention provides a computer-readable storage medium.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps in the method of extracting a street tree from an onboard laser point cloud as described in the first aspect above.
A fourth aspect of the invention provides a computer device.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the method of extracting a street tree from an in-vehicle laser point cloud as described in the first aspect above when the program is executed.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the classification and identification of the street tree point cloud are carried out by utilizing the vehicle-mounted laser point cloud data and a morphological method, and meanwhile, the accurate and rapid identification of the street tree is realized based on the rapid statistics of the information such as the tree diameter, the crown width and the like of the street tree point cloud.
The invention adopts a filtering algorithm to realize the separation of the ground points and the non-ground points; filtering laser points with large coordinate dispersion by adopting a three-dimensional network filtering algorithm, and extracting tree crown point clouds; the improvement is carried out from the two aspects, so that the speed and the accuracy of channel tree extraction are improved, and the effect is better.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
FIG. 1 is a flow chart illustrating a method of extracting a street tree from an onboard laser point cloud, in accordance with the present invention;
FIG. 2 is a schematic view of a neighborhood 3×3 lattice according to the present invention;
fig. 3 is a schematic diagram of a grid in accordance with the present invention.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
It is noted that the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present disclosure. It should be noted that each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by special purpose hardware-based systems which perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.
Example 1
As shown in fig. 1, the present embodiment provides a method for extracting a street tree from a vehicle-mounted laser point cloud, and the present embodiment is illustrated by applying the method to a server, and it can be understood that the method may also be applied to a terminal, and may also be applied to a system and a terminal, and implemented through interaction between the terminal and the server. The server can be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, and can also be a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network servers, cloud communication, middleware services, domain name services, security services CDNs, basic cloud computing services such as big data and artificial intelligent platforms and the like. The terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited herein. In this embodiment, the method includes the steps of:
acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data;
performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data;
dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value;
and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
Specifically, the following may be referred to for a specific implementation procedure of this embodiment:
1. the characteristic that the z values of the laser scanning coordinates of adjacent ground points are similar is utilized, and the separation of the ground points and the non-ground points is realized by using a z value filtering algorithm of a local neighborhood 3X 3 lattice.
The most obvious feature of the ground point compared to other ground features is the lowest point in a range that is approximately horizontal, with both features, the following algorithm was designed by this embodiment.
In this embodiment, first, the characteristic that the z values of local laser scanning coordinates (relative to the height value of the scanner) of adjacent ground points are similar is utilized, and a z value filtering algorithm of a 3×3 lattice (as shown in fig. 2) is used to implement coarse detection of the ground points. The specific algorithm is as follows:
(1-1) taking the local laser scanning coordinate z value in the whole point cloud to be smaller than the parameter z g Constitute a point set P;
setting any laser point in P as the point, taking two adjacent left and right points on the same scanning line with the point, the point closest to the point on the previous scanning line and the next scanning line of the point and the two adjacent left and right points respectively to form a 3X 3 lattice;
(1-3) calculating a Z-value convolution sum Σz of the lattice using the following formula;
wherein Z is 0 The Z value, Z, of the laser scanning coordinate system of the point 1 ~Z 8 The z-values of the coordinate system are scanned for the lasers of adjacent points.
(1-4) setting a threshold Z to ΣZ sum When the dot matrix is smaller than the threshold value, marking all the dots in the dot matrix;
(1-5) traversing all laser points in P, skipping points that have been marked without further processing, and repeating the above operation without marking. The marked points are ground point cloud data, the unmarked points are non-ground point cloud data, and the calculation time can be saved by not repeating the operation on the marked points.
2. Aiming at the characteristic that the rod is generally cylindrical in the rod part, a projection weighing method is adopted to count the laser point information of each projection square, a RANSAC algorithm is utilized to carry out cylinder fitting, and the rod point cloud is extracted.
According to the non-ground point cloud obtained in the last step, calculating the extremum of X, Y coordinates of the non-ground point: x is X Max 、X Min 、Y Max 、Y Min Then traversing point cloud data, constructing a row×column grid with a length according to the extreme value of the point cloud, and then respectively storing the point cloud data into a two-dimensional grid, wherein the Row and Column numbers Row and Column of the grid are calculated as follows:
in the method, in the process of the invention,
length is the grid side length, which is a round-up function.
To facilitate subsequent use of the point cloud within the mesh, any of the points is calculated
（/>
) Line number of point->
Column number->
As shown in the following formula:
in the method, in the process of the invention,
length is the grid side length, which is a function of rounding down.
Any is calculated using the following formula
（/>
) Grid number of the point:
the point cloud gridding schematic diagram is shown in fig. 3, after gridding the point cloud, selecting a grid with the grid density larger than a preset grid density threshold, and taking all points in the grid, namely a plane projection weighing method.
And (3) performing cylinder fitting on the points in the candidate grids by using a RANSAC algorithm, and outputting all point clouds on the cylinder which is successfully fitted. The specific method comprises the following steps:
(2-1) randomly selecting a subset S from the data set S obtained after projection weighing, wherein S is an assumed inner point (the subset S is generally a minimum subset, for example, a straight line selects two points and a circle selects three points);
(2-2) estimating a circular fitting parameter from the subset s;
(2-3) traversing all data in the dataset S except for the subset S, marking as an interior point if the data point is within a given error e, otherwise marking as an exterior point;
(2-4) forming a consistent set by all the interior points, re-estimating the model parameters by using all the interior points in the consistent set if the number of the consistent set points meets a given threshold T, and ending the algorithm;
(2-5) if the number of inliers in the consistent set is less than a threshold T, reselecting a new subset s and repeating (2-1) - (2-4);
and (2-6) after K iterations, selecting a consistent set with the largest number of interior points, re-estimating the model parameters by using all the interior points in the consistent set, and ending the algorithm.
And finally, respectively outputting the identified cylinders through RANSAC fitting of the point cloud data to obtain rod-shaped object point clouds, and providing seed points for clustering operation.
3. By utilizing the characteristic of irregular space morphology of the tree crown point cloud and using a local neighborhood 3 multiplied by 3 stereo network filtering algorithm, filtering laser points with large coordinate dispersion, and extracting tree crown point clouds;
in the vehicle-mounted point cloud data, the tree crowns are the most irregular ground object types, and the three-dimensional network filtering algorithm is designed by utilizing the irregularity of the tree crowns to realize classification of the tree crown point cloud. The specific algorithm flow is as follows:
(3-1) firstly, dividing the unlabeled residual point cloud into three-dimensional grids, wherein the grid size is usually set to be 1 meter, marking each grid, and the initial marking value is 0;
(3-2) traversing the grids one by one, and marking as 1 if the laser points exist in the current grids;
(3-3) statistics of 3X 3 around each grid the number of lattices NG marked 1 in the range;
(3-4) if NG > threshold TN, marking all points within the lattice as crown points;
and (3-5) classifying crown points on a specific trunk according to the trunk position to form an independent tree.
4. And clustering and dividing the crown and the rod-shaped object point cloud respectively, and judging the relative position relationship to obtain the street tree single point cloud.
In this embodiment, the tree crowns and the rod-shaped object point clouds are clustered by using a DBSCAN algorithm, which is a density-based clustering algorithm, and the density clustering algorithm generally assumes that the category can be determined by the compactness of the sample distribution. Samples of the same class are closely connected, that is, there must be samples of the same class around any sample of the class. By grouping closely connected samples into one class, a cluster class is thus obtained. And dividing all the closely connected samples into different categories to obtain the final clustering category results. The specific implementation method is as follows:
(4-1) initializing a core object set seed_queue, initializing a cluster number DB Cluster =0, initializing the set of unvisited samples to be closed_in, for any point P in closed_in i Traversing, and judging whether the object is a core object;
(4-2) find core objects. By cycling through each data point, P is determined i Domain radius r of (2) Eps Whether the number of point clouds in the image is greater than or equal to a density threshold Min Pts If it is, P i Adding a core object set seed_queue for a core object, and marking as a noise point if Pi is not the core object;
(4-3) if the core object set seed_queue is empty, ending the algorithm, otherwise, turning to the step (4-4);
(4-4) randomly selecting one core object Q among the core object set seed_queue i For "seed point", initialize the current cluster core object queue seed_queue= { Q i Initializing cluster number DB Cluster =DB Cluster +1, initializing a current cluster set, and updating unlabeled point cloud closed_in=closed_in-P i ；
(4-5) if the current cluster core object queue seed_queue is an empty set, i.e., all core objects are accessed, the current cluster DB Cluster And (3) after the generation is finished, updating the point cloud cluster, updating the core object set, and turning to the step (3-3). Otherwise, the core object set seed_queue is updated.
(4-6) clustering. A core object, i.e. the current "seed point", is fetched from the current core object queue seed_queue, starting from the field radius r Eps Finding out all density reachable point clouds to generate corresponding cluster, and marking the cluster as the same cluster DB Cluster Updating the current cluster set, updating unlabeled point clouds, and turning to the step (4-5);
and the output result is n point cloud cluster clusters. The segmentation of the rod-shaped point cloud can be realized through DBSCAN clustering.
The trunk is characterized by the crowns connected to it, which are a series of spatially scattered points, so that x_Length, y_Length and z_Length of the crown point block are all larger, where x_Length, y_Length and z_Length are the ranges of a crown cluster point set in x, y and z directions, respectively. And the plane point position coordinates of the trunk are in the plane range of the current crown.
After DBSCAN clustering, the single point cloud of the street tree can be obtained through the combination of the relative positions of the trunk and the crown.
5. And analyzing the single point cloud of each street tree, obtaining a tree diameter value according to the fitted circle of the trunk point cloud, and calculating a crown value according to the range of the crown point cloud.
(5-1) tree diameter
In general, in business application, the diameter of a trunk at a certain height from the ground is taken as a tree diameter, and the trunk is taken as a ground diameter at a position 0.3 m from the ground, a breast diameter at a position 1.0 m, and a breast diameter at a position 1.3 m.
And intercepting the point cloud with a certain thickness at a specific distance, and obtaining the diameter, namely the tree diameter, through circle fitting.
(5-2) crown webs
The average value of the width of the crown in the north-south direction and the east-west direction is used as a reference standard for measuring the growth vigor of the seedlings, and is generally used for representing the specifications of the trees and the seedlings. The coordinate range values x_Length and y_Length of each crown point block can be obtained from the previous step, and the crown value of each tree can be obtained by using the formula (x_Length+y_Length)/2.
In this embodiment, the accuracy of the extraction result of the rod street tree is evaluated by adopting two indexes, namely the integrity rate and the accuracy, and the integrity rate and the accuracy are obtained by taking a single street tree as a basic unit and counting the number, as shown in table 1.
The integrity rate refers to the proportion of the number of correctly extracted ground objects to the total number of the ground objects in the point cloud of the experimental area, and reflects the extracted integrity; the accuracy refers to the proportion of the number of correctly extracted features to the total number of the features extracted, reflecting the accuracy of the extraction.
Table 1 test results
Therefore, the method of the embodiment has higher integrity and accuracy of extracting the pavement tree, and can be applied to batch engineering.
Example two
The embodiment provides a system for extracting a street tree from an on-vehicle laser point cloud.
A system for extracting a street tree from an onboard laser point cloud, comprising:
a data acquisition and separation module configured to: acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data;
a shaft extraction module configured to: performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data;
a crown extraction module configured to: dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value;
an output module configured to: and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
A computing module configured to: crown values are calculated according to the crown point cloud data, and tree diameter values are calculated according to the shaft point cloud data.
It should be noted that the data acquisition and separation module, the shaft extraction module, the crown extraction module, the output module, and the calculation module are the same as the examples and application scenarios implemented by the steps in the first embodiment, but are not limited to the disclosure in the first embodiment. It should be noted that the modules described above may be implemented as part of a system in a computer system, such as a set of computer-executable instructions.
Example III
The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps in the method of extracting a street tree from an on-board laser point cloud as described in the above embodiment.
Example IV
The present embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the steps in the method for extracting a street tree from an on-vehicle laser point cloud according to the above embodiment when executing the program.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, magnetic disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random access Memory (Random AccessMemory, RAM), or the like.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method of extracting a street tree from a vehicle-mounted laser point cloud, comprising:
acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data;
performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data;
dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value;
and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
2. The method of extracting a street tree from an onboard laser point cloud of claim 1, wherein the filtering the onboard laser point cloud data comprises:
taking points with the z value of the local laser scanning coordinate smaller than a set parameter in vehicle-mounted laser point cloud data to form a point set P;
setting any laser point in P as the point, taking two adjacent left and right points on the same scanning line with the point, the point closest to the point on the previous scanning line and the next scanning line of the point and the two adjacent left and right points respectively to form a 3X 3 lattice;
calculating the Z-value convolution sum sigma Z of the 3X 3 lattice;
if sigma Z is smaller than the set threshold Z sum Marking all points in the dot matrix;
traversing all laser points in P, skipping points which are marked and not processing any more, and repeating the processes of constructing a dot matrix, calculating a z-value convolution sum and marking if the points are not marked;
the marked points in the vehicle-mounted laser point cloud data are ground point cloud data, and the unmarked points are non-ground point cloud data.
3. The method for extracting a street tree from an on-vehicle laser point cloud as claimed in claim 1, wherein said process of meshing non-ground point cloud data using a projective weighing method comprises:
calculating the extreme value of the abscissa and the ordinate of the non-ground point cloud data;
constructing a grid according to the extreme values of the abscissa and the ordinate of the non-ground point cloud data, and storing the non-ground point cloud data into the grid;
and selecting a grid with the grid density larger than a preset grid density threshold by adopting a plane projection weighing method.
4. A method of extracting a street tree from an on-board laser point cloud as claimed in claim 3, wherein fitting the point cloud data within the mesh forming the cylinder, extracting the rod point cloud data comprises:
randomly selecting a subset S from a data set S obtained after projection weighing, wherein S is an assumed interior point;
estimating model parameters according to the subset s;
traversing all data except subset S in data set S, marking as an interior point if the data point is within a given error e, otherwise marking as an exterior point;
if the number of the consistent concentrated points meets a given threshold T, re-estimating model parameters by using all the internal points in the consistent concentrated points, and ending the algorithm;
if the number of the inner points in the consistent set is less than the threshold value T, reselecting a new subset s, and reselecting the subset s, estimating model parameters, marking and judging the threshold value;
after K iterations, selecting a consistent set with the largest number of internal points, re-estimating model parameters by using all the internal points in the consistent set, and ending the algorithm;
and obtaining the identified cylinder through RANSAC fitting of the point cloud data.
5. The method of extracting a street tree from an onboard laser point cloud of claim 1, wherein the process of extracting tree crown point cloud data comprises:
dividing three-dimensional grids for non-ground point cloud data;
traversing grids one by one, and marking as 1 if laser points exist in the current grids;
counting the number NG of grids marked as 1 in the range of 3 multiplied by 3 around each grid;
if NG > threshold TN, marking all points in the grid as tree crown points;
and classifying crown points on the specific trunk according to the trunk position to form an independent tree.
6. The method of extracting a street tree from an onboard laser point cloud of claim 1, wherein the process of obtaining street tree single point cloud data comprises:
clustering the crown point cloud data and the rod-shaped object point cloud data by adopting a DBSCAN algorithm, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
7. The method of extracting a street tree from an in-vehicle laser point cloud of claim 1, further comprising: crown values are calculated according to the crown point cloud data, and tree diameter values are calculated according to the shaft point cloud data.
8. A system for extracting a street tree from a vehicle-mounted laser point cloud, comprising:
a data acquisition and separation module configured to: acquiring vehicle-mounted laser point cloud data, and filtering the vehicle-mounted laser point cloud data to separate ground point cloud data and non-ground point cloud data;
a shaft extraction module configured to: performing grid-meshing on non-ground point cloud data by adopting a projection weighing method, fitting the point cloud data in the grid forming a cylinder, and extracting rod-shaped object point cloud data;
a crown extraction module configured to: dividing three-dimensional grids for non-ground point cloud data, counting the number of grids with laser points in the grids, and extracting tree crown point cloud data when the number of grids is larger than a set threshold value;
an output module configured to: and respectively dividing the rod-shaped object point cloud data and the crown point cloud data, and combining the relative position relationship between the trunk and the crown to obtain the street tree single point cloud data.
9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps in the method of extracting a street tree from an on-board laser point cloud as claimed in any of claims 1-7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the program, performs the steps in the method of extracting a street tree from an in-vehicle laser point cloud as claimed in any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310409469.XA CN116129391B (en) | 2023-04-18 | 2023-04-18 | Method and system for extracting pavement tree from vehicle-mounted laser point cloud |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310409469.XA CN116129391B (en) | 2023-04-18 | 2023-04-18 | Method and system for extracting pavement tree from vehicle-mounted laser point cloud |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116129391A true CN116129391A (en) | 2023-05-16 |
| CN116129391B CN116129391B (en) | 2023-07-14 |
Family
ID=86299519
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310409469.XA Active CN116129391B (en) | 2023-04-18 | 2023-04-18 | Method and system for extracting pavement tree from vehicle-mounted laser point cloud |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116129391B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100250482A1 (en) * | 2009-03-24 | 2010-09-30 | Weyerhaeuser Nr Company | System and method for identifying trees using lidar tree models |
| CN106918311A (en) * | 2017-01-20 | 2017-07-04 | 山东科技大学 | Isolated tree crown mapping area automatic calculating method based on vehicle-mounted laser cloud data |
| CN111142116A (en) * | 2019-09-27 | 2020-05-12 | 广东亿嘉和科技有限公司 | Road detection and modeling method based on three-dimensional laser |
| CN112132840A (en) * | 2020-09-01 | 2020-12-25 | 济南市房产测绘研究院(济南市房屋安全检测鉴定中心) | Vehicle-mounted street tree point cloud classification and characteristic information extraction method |
| CN114119863A (en) * | 2021-11-09 | 2022-03-01 | 南京市测绘勘察研究院股份有限公司 | Method for automatically extracting street tree target and forest attribute thereof based on vehicle-mounted laser radar data |
| CN114241155A (en) * | 2021-12-06 | 2022-03-25 | 电子科技大学长三角研究院(湖州) | Urban tree three-dimensional visualization method based on vehicle-mounted laser point cloud data |
-
2023
- 2023-04-18 CN CN202310409469.XA patent/CN116129391B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100250482A1 (en) * | 2009-03-24 | 2010-09-30 | Weyerhaeuser Nr Company | System and method for identifying trees using lidar tree models |
| CN106918311A (en) * | 2017-01-20 | 2017-07-04 | 山东科技大学 | Isolated tree crown mapping area automatic calculating method based on vehicle-mounted laser cloud data |
| CN111142116A (en) * | 2019-09-27 | 2020-05-12 | 广东亿嘉和科技有限公司 | Road detection and modeling method based on three-dimensional laser |
| CN112132840A (en) * | 2020-09-01 | 2020-12-25 | 济南市房产测绘研究院(济南市房屋安全检测鉴定中心) | Vehicle-mounted street tree point cloud classification and characteristic information extraction method |
| CN114119863A (en) * | 2021-11-09 | 2022-03-01 | 南京市测绘勘察研究院股份有限公司 | Method for automatically extracting street tree target and forest attribute thereof based on vehicle-mounted laser radar data |
| CN114241155A (en) * | 2021-12-06 | 2022-03-25 | 电子科技大学长三角研究院(湖州) | Urban tree three-dimensional visualization method based on vehicle-mounted laser point cloud data |
Non-Patent Citations (2)
| Title |
|---|
| JURAJ ČERŇAVA ET AL.: "Processing Chain for Estimation of Tree Diameter from GNSS-IMU-Based Mobile Laser Scanning Data", 《REMOTE SENSING》 * |
| 吴杭彬: "基于林地激光点云的树木胸径自动提取方法", 《同济大学学报(自然科学版)》, pages 947 - 954 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116129391B (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106529469B (en) | Unmanned aerial vehicle-mounted LiDAR point cloud filtering method based on self-adaptive gradient | |
| CN105046235B (en) | The identification modeling method and device of lane line, recognition methods and device | |
| CN109829399A (en) | A kind of vehicle mounted road scene point cloud automatic classification method based on deep learning | |
| CN111598823A (en) | Multi-source mobile measurement point cloud data air-ground integrated fusion method and storage medium | |
| CN111291776B (en) | Channel information extraction method based on crowd source track data | |
| CN110781756A (en) | Urban road extraction method and device based on remote sensing image | |
| CN111192284A (en) | Vehicle-mounted laser point cloud segmentation method and system | |
| CN100595782C (en) | Classification method for syncretizing optical spectrum information and multi-point simulation space information | |
| CN109979008A (en) | A kind of point cloud band division methods based on attribute | |
| CN111754618B (en) | Object-oriented live-action three-dimensional model multi-level interpretation method and system | |
| CN110827302A (en) | Point cloud target extraction method and device based on depth map convolutional network | |
| CN110969142A (en) | Abnormal driving scene extraction method based on internet vehicle natural driving data | |
| CN115457212A (en) | Tree image processing method and device, terminal equipment and storage medium | |
| CN115063555A (en) | Method for extracting vehicle-mounted LiDAR point cloud street tree growing in Gaussian distribution area | |
| CN112053371A (en) | Water body extraction method and device in remote sensing image | |
| CN105740878B (en) | The Symbol Recognition and device of GIS-Geographic Information System | |
| CN114241155A (en) | Urban tree three-dimensional visualization method based on vehicle-mounted laser point cloud data | |
| Zheng et al. | Single shot multibox detector for urban plantation single tree detection and location with high-resolution remote sensing imagery | |
| CN116129391B (en) | Method and system for extracting pavement tree from vehicle-mounted laser point cloud | |
| CN111598027B (en) | Regional shallow water planting area distribution system and method based on remote sensing big data | |
| CN117197677A (en) | Tropical rain forest arbor-shrub separation method based on laser radar point cloud data | |
| CN104951752A (en) | Method for extracting houses from airborne laser point cloud data | |
| CN116993555A (en) | Partition method, system and storage medium for identifying territory space planning key region | |
| CN108399413B (en) | Picture shooting area identification and geographical positioning method and device | |
| CN115311576A (en) | Natural village boundary extraction method based on rural house distribution |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |