CN113222917B - DBI tree vertex detection method of airborne laser radar point cloud data CHM - Google Patents

DBI tree vertex detection method of airborne laser radar point cloud data CHM Download PDF

Info

Publication number
CN113222917B
CN113222917B CN202110470600.4A CN202110470600A CN113222917B CN 113222917 B CN113222917 B CN 113222917B CN 202110470600 A CN202110470600 A CN 202110470600A CN 113222917 B CN113222917 B CN 113222917B
Authority
CN
China
Prior art keywords
dbi
canopy
model
foreground
canopy height
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
Application number
CN202110470600.4A
Other languages
Chinese (zh)
Other versions
CN113222917A (en
Inventor
周国清
穆叶煊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guilin University of Technology
Original Assignee
Guilin University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guilin University of Technology filed Critical Guilin University of Technology
Priority to CN202110470600.4A priority Critical patent/CN113222917B/en
Publication of CN113222917A publication Critical patent/CN113222917A/en
Application granted granted Critical
Publication of CN113222917B publication Critical patent/CN113222917B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • 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
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • 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
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • 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
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G06T2207/10044—Radar image
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30188—Vegetation; Agriculture

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

The invention relates to the field of laser radar data processing, discloses a DBI tree vertex detection method of an airborne laser radar point cloud data canopy height model, and discloses a method for controlling foreground pixels to detect tree vertices based on a DBI of the canopy height canopy model, which comprises the following steps: 1. standardizing the generation condition of the marking pixels of the canopy height model; 2. filtering and removing the pseudo foreground pixels by utilizing the gray change characteristic of the height canopy model; 3. and introducing a similarity judgment factor DBI to carry out DBI-K screening on crown vertexes. The invention provides a method for detecting the tree vertex position of a canopy height model of airborne laser radar point cloud data, which can effectively solve the problem of threshold dependence of the traditional method for detecting the tree vertex by a window and improve the accuracy of identifying the tree vertex.

Description

DBI tree vertex detection method of airborne laser radar point cloud data CHM
Technical Field
The invention relates to the field of Laser radar (LiDAR) data processing, in particular to a DBI (Davies-Bouldin Index) tree vertex Detection method of an airborne LiDAR point cloud data Canopy Height Model (CHM).
Technical Field
The laser radar has strong penetrability to the forest canopy, is very favorable for forest general survey and forest stand parameter acquisition, and can accurately identify the treetop position and the single-tree canopy. The airborne laser radar technology provides theoretical basis and technical support for the development of accurate forestry, and has wide application prospect in forest management and ecological system research.
At present, the tree vertex identification in the field of tree crown position identification is inaccurate due to the complexity of generating a canopy height model image by an airborne LiDAR point cloud. In the existing traditional method, researchers frequently use a method of exploring local maximum values by using a fixed window size to detect tree tops, however, the common method of detecting tree tops based on windows is easily affected by different sizes of tree crown diameters, and if the window is too small, trees with large crown radius are distributed with a plurality of tree tops; conversely, if the window size is too large, this will cause some trees with crown radii smaller than the specified window size to be unassigned to tree vertices.
In summary, in order to solve the problems of threshold dependency and inaccurate identification of the conventional method for detecting tree vertices through windows, it is necessary to provide a new method for identifying tree vertices, so as to solve the above problems. According to the method, the foreground mark pixels are generated according to the gray level characteristics of the canopy height model, and the similarity judgment factor-class spacing ratio DBI is introduced, so that the problem of threshold dependence of the traditional method for detecting the tree vertex of the height canopy model by using a window can be solved, and the identification accuracy of the tree crown vertex is improved.
Disclosure of Invention
The invention aims to provide a method for detecting the tree vertex position of a canopy height model of airborne laser radar point cloud data, which can effectively solve the problem of threshold dependence of the traditional method for detecting the tree vertex by a window and improve the accuracy of identifying the tree vertex.
In order to achieve the purpose of the invention, the invention provides a DBI treetop position detection method of airborne laser radar point cloud data CHM, which is realized by adopting the following technical scheme: a method for controlling foreground pixels to generate tree vertexes based on DBI of a canopy height canopy model comprises the steps of standardizing a mark pixel generation condition of the canopy height model, filtering and removing pseudo foreground pixels by utilizing the gray change characteristic of the height canopy model, introducing a similarity judgment factor DBI, and screening the tree vertexes by DBI-K (DBI-Kmeans); the generation condition of the marking pixel of the standard canopy height model is as follows: (1) the gray level values of the pixels outside the mark are lower than those inside the mark, (2) the pixels of the foreground image form a connected component, and (3) the pixels inside the same mark have the same gray level value; filtering and eliminating the pseudo foreground pixels by utilizing the gray level change characteristics of the height canopy model, further processing the foreground mark pixels which are calculated and solved by combining the gray level change characteristics of the height canopy model image, and filtering background points which appear on the foreground mark according to the difference of different image gray levels; and (4) screening crown vertexes by DBI-K (DBI-Kmeans), carrying out DBI-K clustering on the foreground images, obtaining the optimal solution of a clustering center when the DBI is the minimum value, taking the optimal solution as the treetop of the single tree, and stopping the algorithm.
The invention has the beneficial effects that: the method comprises the steps of controlling foreground pixels to generate tree vertexes by using a DBI (digital base interface) based on a canopy height canopy model, including the steps of standardizing mark pixel generation conditions of the canopy height model, filtering and removing pseudo foreground pixels by using the gray change characteristics of the height canopy model, and introducing a similarity judgment factor DBI to carry out DBI-K (DBI-Kmeans) screening of the tree crown vertexes, so that the problem of threshold dependence of a traditional window detection tree vertex method is solved, and the tree vertex identification accuracy is improved; on one hand, the generation conditions of the marking pixels of the canopy height model are normalized, the local maximum value of the image of the canopy height model is utilized to determine the region of interest, and the region of interest is used as the foreground pixels and the marking, so that the problem of threshold value dependence of cross-cut detection can be avoided; on the other hand, the gray level change characteristic of the height canopy model is used for filtering and removing the pseudo foreground pixels, so that the problem that pseudo maximum values (foreground marks) are doped in the foreground marks due to textures and noise is solved; and finally, introducing a similarity judgment factor DBI to carry out DBI-K screening on tree crown vertexes, solving the problem of ambiguity of the tree vertexes based on the canopy height model, and realizing effective identification of the tree vertexes.
The direct feature-based foreground marker pixel extraction results in the adulteration of pseudo-maxima in the foreground marker due to the existence of texture and noise.
Drawings
FIG. 1 is a point cloud data filtering classification diagram.
FIG. 2 is a DEM, DSM, CHM map generated from point cloud data.
Fig. 3 is a schematic diagram of the present invention.
FIG. 4 is a foreground image and foreground label of the canopy height model generated under canonical conditions by the present invention.
FIG. 5 is a diagram of the foreground after filtering and rejecting the pseudo foreground pixels according to the present invention.
FIG. 6 is a DBI-K screening crown top plot of the present invention.
Detailed Description
The following further describes the embodiments of the present invention with reference to the drawings.
The embodiment is as follows:
step 1) filtering and classifying the point cloud data by combining the graph 1. On the premise that the reservation of the topographic information of the forest region sample plot is very important, in order to achieve a better filtering effect, the invention adopts a progressive triangulation network filtering algorithm, and the following steps are taken for separating the ground points:
(1) setting the maximum terrain slope: setting the maximum slope of the terrain displayed in the point cloud to be 80 degrees;
(2) setting an iteration angle: converting angles of all allowable values between the classification points of the ground to be detected and the currently known classification points of the ground into 8 degrees;
(3) setting an iteration distance: setting the distance threshold between the point to be classified and the triangulation network to be 1.6 m;
by using the method, the point cloud filtering classification is carried out on the experimental sample plot, the separated ground point class is displayed in brown, and the separated vegetation point class is displayed in green.
And step 2) generating the DEM, the DSM and the CHM by using the filtered and classified point cloud data in combination with the graph 2. And (3) comprehensively utilizing a TIN calculation method to obtain non-ground laser point clouds and ground reflection laser point clouds, and respectively generating DSM and DEM of the experimental forest area. After multiple experiments in LiDAR 360, it was found that the triangulation generated when the critical lengthening was set to 1m was smoother, the triangulation structure face contained more detail when the value of the insert buffer was set to 2m, and the weight was set to 2. The DEM is subtracted from the DSM to obtain a model of the canopy height, as shown in the following equation:
CHM＝DSM-DEM (1)
step 3) combining with the graph 3, controlling a foreground pixel to generate a tree vertex based on the DBI of the canopy height model, inputting the canopy height model in the generation of the mark pixel of the canopy height model under the standard condition, and generating the foreground mark pixel of the canopy height model according to three standard conditions; in the process of filtering and removing the pseudo foreground pixels by utilizing the gray level change characteristics of the height canopy model, inputting the initial foreground marking pixels of the canopy height model, and filtering and removing the pseudo foreground marking pixels C according to the local maximum gray level pixel principleFront side(ii) a Introducing a similarity judgment factor DBI to carry out DBI-K screening on crown vertexes, inputting foreground mark pixels for filtering pseudo marks, taking the foreground mark pixels as initial clustering centers, generating new clustering centers according to a clustering center formula, comparing values of the similarity judgment factor DBI before and after the clustering centers are generated, and when the DBI is the minimum valueObtaining the optimal solution of the clustering center, and terminating the algorithm; and outputting the clustering center when the DBI takes the minimum value, and taking the clustering center as the top point of the single tree.
And 4) generating the marked pixels of the canopy height model under the standard condition by combining the graph 4. Generating an initial foreground marker of the canopy height model according to 3 normative conditions: (1) the gray level values of the pixels outside the mark are lower than those inside the mark, (2) the pixels of the foreground image form a connected component, and (3) the pixels inside the same mark have the same gray level value; the local maximum value of the image determines the interested area, and the interested area is used as the foreground pixel and the mark.
And step 5) combining with the graph 5, filtering and removing the pseudo foreground pixels by the gray level change characteristic of the height canopy model. The tree canopy region of the high gray level canopy model image is brighter than other regions, the relative height information of the canopy can be known through the color depth in the gray level image, namely, the greater the gray level value, the higher the canopy is, and therefore, the treetop (foreground mark) can be known as a local gray level maximum pixel point; since the gray canopy height model is represented by the gray values of 8 pixels adjacent to the center pixel element. A gray image can be obtained by equalizing the values of R, G, B (where R ═ G ═ B ═ 255 is white and R ═ G ═ B ═ 0 is black); when the sum of the R, G, B component values of the canopy height model foreground pixels takes the maximum value and the R, G, B component values are equal, the flag pixel value takes 0, otherwise the flag pixel value takes 1. Expressed as the following equation:
Figure BDA0003045243950000041
in the formula (I), the compound is shown in the specification,
Figure BDA0003045243950000042
and
Figure BDA0003045243950000043
respectively representing R, G, B component values at pixel points in the initial foreground labeled canopy height model image C. Image C can be converted through the formula to obtain a new image C'As the final foreground marker image.
Step 6) screening crown vertexes by DBI-K (DBI-Kmeans) in combination with the graph 6, and further screening the marked pixels by utilizing the similarity judgment factor on the basis of generating the foreground marked pixels of the height canopy model; clustering is carried out on the input two-dimensional gray level image of the canopy height model, DBI similarity judgment factors are added into a K-means clustering method, and clustering center m and clustering center number K values in the foreground image are further determined.
Assume that a given sample canopy height model foreground image contains a dataset a ═ ai}; and initially dividing n sample data sets in the A into K different clusters, and ensuring that the intra-cluster similarity is higher and the extra-cluster similarity is lower. Therefore, the core theory of the K-means clustering method can be described as: clustering a data set A into a set B consisting of K clusters, wherein B is { B ═ B }i}; the intra-cluster centers of each cluster subset can be found to be:
Figure BDA0003045243950000044
where A represents a sample data object and NiRepresenting a cluster biNumber of samples in (1).
And when the value of the similarity judgment factor DBI is minimum, outputting the DBI similarity judgment factor to obtain the optimal solution of the clustering number K.
Figure BDA0003045243950000045
In the formula, aiRepresenting individual data objects, m, within the ith classiCluster center, N, representing the ith classiIndicates the number of data objects in the ith class, WiIndicates the degree of dispersion, W, of the data objects in the ith classjIndicating the degree of dispersion of each data object in the jth class, Di,jRepresenting the euclidean distance between the centroids of the ith and jth classes.
D in formula (1.2)i,jRepresenting inter-class distance, expressionThe formula is as follows:
Di,j＝||mi-mj|| (5)
w in equation (1.2)iThe intra-class distance is expressed by the following expression formula:
Figure BDA0003045243950000051
if DBInew＜DBIlastA new cluster center may be formed, otherwise the algorithm terminates. And determining tree vertexes and the number K when iteration is finished, taking a clustering center generated by the iteration termination as a treetop point, taking a clustering K value as the number of the single plants of the target image, and taking the screened clustering center pixel as the position of a seed point (treetop point) for describing each crown.
Figure BDA0003045243950000052
In the formula (I), the compound is shown in the specification,&&for the logical relationship ' AND ', m ' is the new set of centers, m1…j… is the clustering center m1To mi，m1...iAs the clustering center m1To mj. When the DBI takes the minimum value, the optimal solution is obtained, and the number and the position of the tree vertexes are determined at the moment.
The above description is only for the purpose of describing the preferred embodiment of the present invention in conjunction with the accompanying drawings, and it should be noted that various changes and modifications within the scope of the appended claims may be made by those skilled in the art, and these changes and modifications should also be construed as being within the scope of the present invention.

Claims (3)

1. The DBI tree vertex detection method of airborne laser radar point cloud data CHM comprises the following specific steps:
step 1) filtering and classifying point cloud data and generating a Digital Elevation Model (DEM), a Digital Surface Model (DSM) and a Canopy Height Model (Canopy Height Model, CHM);
step 2) standardizing the foreground mark pixel generation conditions of the canopy height model according to the pixel gray level of the height canopy model (CHM) to generate an initial foreground mark of the canopy height model, which specifically comprises the following steps:
(1) the gray level values of the pixels outside the canopy height model image markers are all lower than those inside the markers;
(2) forming a connected component by pixel points of the foreground image of the canopy height model image;
(3) the canopy height model image has the same gray level value with the pixel point inside the same mark;
generating a canopy height model image C after the initial foreground marking according to the 3 standard conditions;
step 3) filtering and removing the pseudo foreground pixels by utilizing the gray value change characteristics of the pixels in eight adjacent domains of the central pixel based on the initial foreground mark of the height canopy model;
step 4) introducing a similarity judgment factor DBI, and carrying out crown vertex screening of a DBI-K method on the canopy height model foreground mark, wherein the method specifically comprises the following steps:
clustering the input two-dimensional gray level images of the canopy height model, and generating clustering center m and clustering center number K values of the foreground images of the canopy height model by utilizing a K-means clustering method introducing a similarity judgment factor DBI (direct binary input) for multiple times of iteration; if DBInew＜DBIlastForming a new cluster center, otherwise, terminating the algorithm, and taking the cluster center m generated by iteration termination as a treetop point TpointThe value K of the number of the clustering centers is used as the number of the single plants of the target image, and the screened clustering center pixels are used for describing the treetop point T of each crownpointThen the tree vertices are labeled and displayed visually in three dimensions.
2. The method according to claim 1, characterized in that said step 1) is in particular:
in order to ensure the integrity of terrain confidence and the filtering and classifying effect of laser radar point cloud, the method adopts a progressive triangulation network filtering algorithm, filtering and classifying the experimental sample plot data by setting a maximum gradient, an iteration angle and an iteration distance, performing coloring display on the point cloud after filtering and classifying, calculating the non-ground laser point cloud and the ground reflected laser after filtering and classifying by using a TIN method to obtain DSM and DEM which respectively generate experimental forest zones, and subtracting the DEM from the DSM to obtain CHM.
3. The method according to claim 1, characterized in that said step 3), in particular:
the method utilizes R, G, B values to obtain gray level image of the canopy height model, when C of foreground pixel of the canopy height modelR、CG、CBThe sum of the component values is taken to be the maximum value and the requirement R, G, B that the component values are equal, the pixel value of the foreground pixel is marked to be 0, otherwise the pixel value is 1 in all cases, resulting in a new image C' as the final foreground marked image.
CN202110470600.4A 2021-04-29 2021-04-29 DBI tree vertex detection method of airborne laser radar point cloud data CHM Active CN113222917B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110470600.4A CN113222917B (en) 2021-04-29 2021-04-29 DBI tree vertex detection method of airborne laser radar point cloud data CHM

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110470600.4A CN113222917B (en) 2021-04-29 2021-04-29 DBI tree vertex detection method of airborne laser radar point cloud data CHM

Publications (2)

Publication Number Publication Date
CN113222917A CN113222917A (en) 2021-08-06
CN113222917B true CN113222917B (en) 2022-06-14

Family

ID=77089898

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110470600.4A Active CN113222917B (en) 2021-04-29 2021-04-29 DBI tree vertex detection method of airborne laser radar point cloud data CHM

Country Status (1)

Country Link
CN (1) CN113222917B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106023178A (en) * 2016-05-16 2016-10-12 浙江工业大学 Method for detecting single tree in remote sensing data based on gradient direction clustering
CN106845399A (en) * 2017-01-18 2017-06-13 北京林业大学 A kind of method that use hierarchical cluster mode extracts individual tree information from LiDAR point cloud
CN109164459A (en) * 2018-08-01 2019-01-08 南京林业大学 A kind of method that combination laser radar and high-spectral data classify to forest species
CN110389369A (en) * 2019-07-30 2019-10-29 南京林业大学 Canopy point cloud acquisition methods based on RTK-GPS and mobile two dimensional laser scanning
CN110992375A (en) * 2019-11-29 2020-04-10 桂林理工大学 Chord angle discrimination clustering single tree segmentation method for layer-by-layer LiDAR point cloud
CN111739031A (en) * 2020-06-19 2020-10-02 华南农业大学 Crop canopy segmentation method based on depth information

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AUPR301401A0 (en) * 2001-02-09 2001-03-08 Commonwealth Scientific And Industrial Research Organisation Lidar system and method
CN104820830B (en) * 2015-05-08 2018-01-02 南京林业大学 A kind of wood recognition method based on Full wave shape LiDAR canopy section models
CN107274417B (en) * 2017-07-05 2020-06-16 电子科技大学 Single tree segmentation method based on airborne laser point cloud aggregation relation
CN109902686B (en) * 2019-01-22 2021-01-15 中国科学院植物研究所 Forest single tree parameter extraction method
CN110196432B (en) * 2019-04-28 2021-04-02 湖南工学院 Deciduous forest tree-level parameter determination method based on small-spot airborne radar
CN110223314B (en) * 2019-06-06 2021-09-24 电子科技大学 Single tree segmentation method based on tree crown three-dimensional point cloud distribution
CN110288594B (en) * 2019-07-02 2021-06-04 河北农业大学 Plant canopy structure character analysis method
CN111428784B (en) * 2020-03-23 2024-06-18 湖南工学院 Robust segmentation method for determining deciduous forest tree level parameters by using airborne laser radar
CN111462134A (en) * 2020-03-31 2020-07-28 武汉大学 Single-tree segmentation method and system for fusing high-resolution remote sensing image and laser radar point cloud
CN112163458A (en) * 2020-09-04 2021-01-01 江苏东晟辉科技开发有限公司 Ground feature classification method based on integration of CASI hyperspectrum and airborne LiDAR
CN112669333B (en) * 2021-01-11 2024-07-12 四川测绘地理信息局测绘技术服务中心 Single wood information extraction method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106023178A (en) * 2016-05-16 2016-10-12 浙江工业大学 Method for detecting single tree in remote sensing data based on gradient direction clustering
CN106845399A (en) * 2017-01-18 2017-06-13 北京林业大学 A kind of method that use hierarchical cluster mode extracts individual tree information from LiDAR point cloud
CN109164459A (en) * 2018-08-01 2019-01-08 南京林业大学 A kind of method that combination laser radar and high-spectral data classify to forest species
CN110389369A (en) * 2019-07-30 2019-10-29 南京林业大学 Canopy point cloud acquisition methods based on RTK-GPS and mobile two dimensional laser scanning
CN110992375A (en) * 2019-11-29 2020-04-10 桂林理工大学 Chord angle discrimination clustering single tree segmentation method for layer-by-layer LiDAR point cloud
CN111739031A (en) * 2020-06-19 2020-10-02 华南农业大学 Crop canopy segmentation method based on depth information

Also Published As

Publication number Publication date
CN113222917A (en) 2021-08-06

Similar Documents

Publication Publication Date Title
Yang et al. An individual tree segmentation method based on watershed algorithm and three-dimensional spatial distribution analysis from airborne LiDAR point clouds
CN113592882B (en) Crown extraction method based on multi-source remote sensing of unmanned aerial vehicle
CN106199557B (en) A kind of airborne laser radar data vegetation extracting method
Wen et al. A novel automatic change detection method for urban high-resolution remotely sensed imagery based on multiindex scene representation
Wang et al. Urban forest monitoring based on multiple features at the single tree scale by UAV
Ramiya et al. Segmentation based building detection approach from LiDAR point cloud
CN105631892B (en) It is a kind of that detection method is damaged based on the aviation image building of shade and textural characteristics
CN108241871A (en) Laser point cloud and visual fusion data classification method based on multiple features
CN106909902A (en) A kind of remote sensing target detection method based on the notable model of improved stratification
CN109164459A (en) A kind of method that combination laser radar and high-spectral data classify to forest species
CN111598048A (en) Urban village-in-village identification method integrating high-resolution remote sensing image and street view image
CN103927511A (en) Image identification method based on difference feature description
CN113963222A (en) High-resolution remote sensing image change detection method based on multi-strategy combination
Wang et al. A novel multi-scale segmentation algorithm for high resolution remote sensing images based on wavelet transform and improved JSEG algorithm
Chen et al. The mixed kernel function SVM-based point cloud classification
CN110348478B (en) Method for extracting trees in outdoor point cloud scene based on shape classification and combination
CN115147746B (en) Saline-alkali geological identification method based on unmanned aerial vehicle remote sensing image
CN112669363A (en) Urban green land three-dimensional green volume calculation method
CN116704333B (en) Single tree detection method based on laser point cloud data
CN115063555A (en) Method for extracting vehicle-mounted LiDAR point cloud street tree growing in Gaussian distribution area
CN111487643A (en) Building detection method based on laser radar point cloud and near-infrared image
CN111860359A (en) Point cloud classification method based on improved random forest algorithm
Al-Ghrairi et al. Classification of satellite images based on color features using remote sensing
Zhou et al. A geographic object-based approach for land classification using LiDAR elevation and intensity
CN113724400B (en) Multi-attribute fusion building point cloud extraction method for oblique photography

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