CN111091030A - Tree species identification method and device, computer equipment and readable storage medium - Google Patents
Tree species identification method and device, computer equipment and readable storage medium Download PDFInfo
- Publication number
- CN111091030A CN111091030A CN201811243826.5A CN201811243826A CN111091030A CN 111091030 A CN111091030 A CN 111091030A CN 201811243826 A CN201811243826 A CN 201811243826A CN 111091030 A CN111091030 A CN 111091030A
- Authority
- CN
- China
- Prior art keywords
- tree species
- distribution map
- forest
- forest region
- acquiring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 97
- 230000011218 segmentation Effects 0.000 claims description 30
- 238000012549 training Methods 0.000 claims description 27
- 238000003066 decision tree Methods 0.000 claims description 22
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 15
- 238000003672 processing method Methods 0.000 claims description 14
- 238000012937 correction Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000007637 random forest analysis Methods 0.000 claims description 8
- 238000007635 classification algorithm Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000007500 overflow downdraw method Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 abstract description 28
- 241000894007 species Species 0.000 description 174
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 19
- 235000017491 Bambusa tulda Nutrition 0.000 description 19
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 19
- 241001330002 Bambuseae Species 0.000 description 18
- 241000219927 Eucalyptus Species 0.000 description 18
- 239000011425 bamboo Substances 0.000 description 18
- 230000000694 effects Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000012545 processing Methods 0.000 description 10
- 230000008859 change Effects 0.000 description 5
- 241000196324 Embryophyta Species 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012952 Resampling Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241000209128 Bambusa Species 0.000 description 1
- 244000018436 Coriandrum sativum Species 0.000 description 1
- 235000002787 Coriandrum sativum Nutrition 0.000 description 1
- 241000018650 Pinus massoniana Species 0.000 description 1
- 235000011609 Pinus massoniana Nutrition 0.000 description 1
- 241000219492 Quercus Species 0.000 description 1
- 235000016976 Quercus macrolepis Nutrition 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a tree species identification method and device, computer equipment and a readable storage medium. The method can comprise the following steps: acquiring laser radar data and multispectral images of a forest area; acquiring a canopy height model of a forest region according to laser radar data; acquiring a vector distribution map of tree species in a forest region according to the canopy height model and preset classification parameters; and obtaining the tree species distribution map according to the vector distribution map and the multispectral image. According to the method, the space three-dimensional coordinates of the ground object points can be directly obtained through the laser radar data, the space form and the geometric relation of the forest region are intuitively and accurately restored, and the precision of the vector distribution diagram of the tree species in the obtained forest region is improved; in addition, the multispectral image can obtain color information of tree species in the forest region, the accuracy of a computer device for obtaining a tree species distribution diagram according to the vector distribution diagram and the multispectral image is improved, and the accuracy of tree species identification and classification in the forest region is further improved.
Description
Technical Field
The invention relates to the field of computer software, in particular to a tree species identification method, a tree species identification device, computer equipment and a readable storage medium.
Background
Each plant on earth usually has many individuals, which are distributed over a certain area. The method for acquiring the planting conditions and the spatial distribution conditions of various plants has important significance for accurately mastering the resource reserves of various plants, so that the tree species in a plant distribution area needs to be identified. For example, eucalyptus and bamboo have the advantages of high growth speed, wide application, good economic benefit and the like, and the control of the planting condition and the spatial distribution condition of the two tree species has important significance for accurately controlling the resource reserves of eucalyptus and bamboo.
Along with the development of aviation laser radar system technology, especially unmanned aerial vehicle laser radar technology, the space form and the geometric relation in vegetation coverage area can be directly perceived accurately to the laser radar data that provide. At present, the method for identifying the tree species uses an object-oriented method, and the data source is data obtained by combining airborne laser radar data and a high-resolution image, a near-infrared image or a high-resolution color near-infrared remote sensing image, so as to identify the tree species.
However, the conventional tree species identification method is difficult to classify tree species, so that the accuracy of identifying tree species is low.
Disclosure of Invention
Based on this, it is necessary to provide a tree species identification method, apparatus, computer device and readable storage medium for solving the problem that the conventional tree species identification method is difficult to classify tree species and the accuracy of identifying tree species is low.
In a first aspect, an embodiment of the present invention provides a method for identifying tree species, where the method includes:
acquiring laser radar data and multispectral images of a forest area;
acquiring a canopy height model of the forest region according to the laser radar data;
acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
In one embodiment, the obtaining a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters includes:
performing single-tree segmentation on the canopy height model by adopting a preset multi-scale segmentation method to obtain a segmented image;
and acquiring a vector distribution map of the tree species in the forest region according to the segmentation image and the preset classification parameters.
In one embodiment, the obtaining a vector distribution map of tree species in the forest region according to the segmented image and the preset classification parameter includes:
determining the classification parameters according to a random forest classification method; the classification parameters comprise the depth of the decision tree and the number of the decision tree;
determining a training sample set according to the segmented image; the training sample set comprises images of various trees in the forest region;
and classifying the segmented images according to the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region.
In one embodiment, the classifying the segmented image according to the training sample set and the classification parameter to obtain a vector distribution map of tree species in the forest region includes:
determining target characteristics according to a preset characteristic selection algorithm and data characteristics; the data features comprise point cloud features corresponding to the laser radar data and image features corresponding to the multispectral image;
and classifying the segmented images according to the target features, the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region.
In one embodiment, the determining the target feature according to the preset feature selection algorithm and the data feature includes:
sorting the data characteristics of the target tree species according to the importance of the data characteristics;
respectively acquiring the first N data characteristics of each target tree species;
and determining the union of the first N data characteristics of each target tree species as the target characteristics.
In one embodiment, the obtaining a canopy height model of the forest area according to the lidar data includes:
acquiring a digital elevation model and a digital surface model according to the laser radar data;
and acquiring the canopy height model according to the digital elevation model and the digital surface model.
According to the tree species identification method provided by the embodiment, computer equipment needs to acquire laser radar data and multispectral images of a forest region, the laser radar data can be acquired by an airborne laser radar system, the multispectral images can be acquired by a remote sensing satellite, the computer equipment acquires a canopy height model of the forest region according to the acquired laser radar data, acquires a vector distribution diagram of tree species in the forest region according to the canopy height model and preset classification parameters of the forest region, and acquires the tree species distribution diagram according to the vector distribution diagram and the multispectral images. According to the method, the laser radar data of the forest region acquired by the airborne laser radar system can directly acquire the spatial three-dimensional coordinates of the ground feature points, the spatial form and the geometric relation of the forest region can be intuitively and accurately restored, the accuracy of a canopy height model of the forest region acquired by computer equipment according to the acquired laser radar data is improved, and therefore the precision of a vector distribution diagram of tree species in the forest region acquired according to the canopy height model and preset classification parameters is improved; in addition, the remote sensing satellite acquires the multispectral image of the forest region, so that the color information of the tree species in the forest region can be obtained, the precision of the computer equipment for obtaining the tree species distribution diagram according to the vector distribution diagram and the multispectral image is improved, and the precision of identifying and classifying the tree species in the forest region is further improved.
In a second aspect, an embodiment of the present invention provides an apparatus for identifying tree species, where the apparatus includes:
the first acquisition module is used for acquiring laser radar data and multispectral images of a forest area;
the second acquisition module is used for acquiring a canopy height model of the forest region according to the laser radar data;
the third acquisition module is used for acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
the determining module is used for obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the following steps when executing the computer program:
acquiring laser radar data and multispectral images of a forest area;
acquiring a canopy height model of the forest region according to the laser radar data;
acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the following steps:
acquiring laser radar data and multispectral images of a forest area;
acquiring a canopy height model of the forest region according to the laser radar data;
acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
The identification device for the tree species, the computer device and the readable storage medium provided by this embodiment enable the computer device to obtain a canopy height model of a forest region according to the obtained lidar data, obtain a vector distribution map of the tree species in the forest region according to the canopy height model of the forest region and preset classification parameters, and obtain the tree species distribution map according to the vector distribution map and a multispectral image. The space three-dimensional coordinates of ground object points can be directly obtained through laser radar data of the forest region acquired by the airborne laser radar system, the space form and the geometric relation of the forest region are intuitively and accurately restored, the accuracy of a canopy height model of the forest region acquired by computer equipment according to the acquired laser radar data is improved, and the precision of a vector distribution diagram of tree seeds in the forest region acquired according to the canopy height model and preset classification parameters is improved; in addition, the remote sensing satellite acquires the multispectral image of the forest region, so that the color information of the tree species in the forest region can be obtained, the precision of the computer equipment for obtaining the tree species distribution diagram according to the vector distribution diagram and the multispectral image is improved, and the precision of identifying and classifying the tree species in the forest region is further improved.
Drawings
FIG. 1 is a schematic diagram of an internal structure of a computer device according to an embodiment;
FIG. 2 is a flowchart illustrating a tree species identification method according to an embodiment;
FIG. 3 is a schematic flow chart illustrating a tree species identification method according to another embodiment;
FIG. 4(a) is a schematic diagram of a segmented image according to an embodiment;
FIG. 4(b) is a schematic diagram of a segmented image according to an embodiment;
FIG. 4(c) is a schematic diagram of a segmented image according to an embodiment;
FIG. 5 is a flowchart illustrating a tree species identification method according to another embodiment;
FIG. 6 is a trend graph of the overall accuracy of classification of tree species in a forest area as a function of decision tree depth provided by one embodiment;
FIG. 7 is a graph illustrating the trend of the overall accuracy of classification and the classification time of tree species in a forest area as a function of the number of decision trees provided in one embodiment;
FIG. 8 is a flowchart illustrating a tree species identification method according to another embodiment;
FIG. 9 is a flowchart illustrating a tree species identification method according to another embodiment;
FIG. 10 is a flowchart illustrating a tree species identification method according to another embodiment;
FIG. 11 is a flowchart illustrating a tree species identification method according to another embodiment;
FIG. 12 is a schematic diagram of an apparatus for identifying tree species according to an embodiment;
FIG. 13 is a schematic diagram of an apparatus for identifying tree species according to an embodiment;
FIG. 14 is a schematic diagram of an apparatus for identifying tree species according to an embodiment;
fig. 15 is a schematic structural diagram of a tree species identification device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The tree species identification method provided by the embodiment of the invention can be applied to computer equipment shown in FIG. 1. The computer device comprises a processor and a memory connected by a system bus, wherein a computer program is stored in the memory, and the steps of the method embodiments described below can be executed when the processor executes the computer program. Optionally, the computer device may further comprise a network interface, a display screen and an input device. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a nonvolatile storage medium storing an operating system and a computer program, and an internal memory. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. Optionally, the computer device may be a server, may be a PC, may also be a personal digital assistant, may also be other terminal devices, such as a PAD, a mobile phone, and the like, and may also be a cloud or a remote server, and the specific form of the computer device is not limited in the embodiment of the present invention.
In the traditional tree species identification method, an object-oriented method is used, the data source is data obtained by combining airborne laser radar data and a high-resolution image, a near-infrared image or a high-resolution color near-infrared remote sensing image, and the tree species are identified. Therefore, the embodiment of the invention provides a tree species identification method, a tree species identification device, a computer device and a readable storage medium, and aims to solve the above technical problems of the conventional technology.
The following describes the technical solution of the present invention and how to solve the above technical problems with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present invention will be described below with reference to the accompanying drawings.
Fig. 2 is a schematic flowchart of a tree species identification method according to an embodiment. The embodiment relates to a specific implementation process for obtaining a tree species distribution map in a forest region. As shown in fig. 2, the method may include:
s201, laser radar data and multispectral images of the forest area are obtained.
Specifically, the laser radar data are obtained according to the time, the signal strength degree, the frequency change and the like of a reflection signal of laser emitted to the forest region by the emitter to the reflector, and the spatial three-dimensional coordinates of ground object points can be directly obtained according to the laser radar data, so that the spatial form and the geometric relation of the forest region can be intuitively and accurately restored; the multispectral image can be obtained through a remote sensing satellite, and the color information of tree species in the forest region can be obtained through the multispectral image. The computer equipment can acquire the laser radar data and multispectral data of the forest region from a laser radar system and a remote sensing satellite.
In the embodiment, the laser radar data of the forest region can be acquired by using an aviation laser radar system, and the laser radar data of the forest region can also be acquired by using an unmanned aerial vehicle laser radar system; the multispectral image of the forest region can be obtained by using a GF-2 optical remote sensing satellite, and the multispectral image of the forest region can also be obtained by using other remote sensing satellites. Or the computer equipment acquires data such as time, signal strength degree, frequency change and the like of a reflection signal of laser emitted to a forest area by the laser radar, and acquires laser radar data according to the acquired data.
S202, acquiring a canopy height model of the forest region according to the laser radar data.
The canopy height model can be used for obtaining forest tree high-level parameters in the forest region. Specifically, the computer device may resample the lidar data to obtain a canopy height model of the forest region. Or, the computer device may further perform preprocessing on the laser radar data, perform resampling processing on the preprocessed laser radar data, and acquire the canopy height model of the forest region, for example, may discard abnormal points, error points, incomplete data points, and the like in the laser radar data, may further perform preprocessing such as repairing abnormal points and missing points in the laser radar data, and then perform resampling processing on the preprocessed laser radar data, and acquire the canopy height model of the forest region.
S203, acquiring a vector distribution map of the tree species in the forest region according to the canopy height model and preset classification parameters.
The classification parameters are used for assisting in obtaining a vector distribution map of tree species in the forest region, and the classification parameters can be verified through the classification precision of the tree species in the forest region to obtain corresponding classification parameters when the classification precision reaches a maximum value, or can be set in advance according to human experience. The vector distribution diagram of the tree species in the forest region can form the outline of the outer frame of the tree species by lines, and the color of the tree species is displayed by the color of the outer frame and the color of the closed region formed by the outer frame. Specifically, the vector distribution maps corresponding to different classification parameters are also different, and the computer device can acquire different vector distribution maps of tree species in the forest region according to the canopy height model and different preset classification parameters. Alternatively, the canopy height model may be segmented to obtain different segmented images, and then a vector distribution map may be obtained according to each segmented image and the classification parameter.
S204, obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
The tree species distribution map can reflect the crown shape, the tree height, the density degree of the tree species and the like of the tree species in the forest region, the types of the tree species in the forest region can be identified according to the tree species distribution map, and the spatial distribution condition of the target tree species is obtained. In this embodiment, the computer device modifies and edits the miscut and missed areas in the vector distribution map of the tree species in the obtained forest area according to the obtained multispectral image, so as to obtain a tree species distribution map for identifying the target tree species in the forest area. For example, the vector distribution map is overlapped with the multispectral image, the miscut and missed area in the vector distribution map of the tree species in the forest area is identified, and the miscut and missed area in the vector distribution map is modified and edited according to the multispectral image to obtain the tree species distribution map.
In this embodiment, the laser radar data of the forest region acquired by the airborne laser radar system can directly acquire the spatial three-dimensional coordinates of the ground object points, and intuitively and accurately restore the spatial form and the geometric relationship of the forest region, so that the accuracy of a canopy height model of the forest region acquired by the computer device according to the acquired laser radar data is improved, and the precision of a vector distribution map of tree species in the forest region acquired according to the canopy height model and preset classification parameters is improved; in addition, the remote sensing satellite acquires the multispectral image of the forest region, so that the color information of the tree species in the forest region can be obtained, the precision of the computer equipment for obtaining the tree species distribution diagram according to the vector distribution diagram and the multispectral image is improved, and the precision of identifying and classifying the tree species in the forest region is further improved.
Fig. 3 is a schematic flowchart of a tree species identification method according to another embodiment. FIG. 4 is a schematic diagram of a segmented image according to an embodiment. The embodiment relates to a specific implementation process for obtaining a vector distribution map of tree species in a forest region according to a canopy height model and preset classification parameters. As shown in fig. 3, on the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S203 includes:
s301, performing single-tree segmentation on the canopy height model by adopting a preset multi-scale segmentation method to obtain a segmented image.
Specifically, the computer device obtains a local highest point in the canopy height model as a seed point by using a grid-based method, and performs single-tree segmentation on the canopy height model by using a multi-scale segmentation method according to the obtained seed point and preset segmentation scale, shape factor and compactness factor to obtain a segmented image, wherein the segmented image is composed of polygons of a plurality of crown boundaries, as shown in fig. 4. When the canopy height models of tree species in the forest region are subjected to single tree segmentation, the canopy height models are subjected to single tree segmentation under different single segmentation scales, so that different segmentation images can be obtained, for example, when the canopy height models of eucalyptus and bamboo in the forest region are subjected to single scale segmentation, fig. 4(a) is a segmentation image obtained when the segmentation scale is 8, the segmentation boundary of the eucalyptus in the segmentation image is consistent with the target size of a ground object, and the segmentation of the bamboo and other ground objects is too broken; fig. 4(b) is a divided image obtained at a division scale of 20, in which eucalyptus is excessively divided and bamboo is underdivided; fig. 4(c) is a divided image obtained when the division scale is 50, in which the matching degree of the bamboo boundary and the decomposition object is high; the segmentation effect of different shape factors and compactness factors is compared through experiments, the segmentation effect when the shape factor is 0.8 is more consistent with the characteristic that the ground object boundary is segmented in a research area, and the boundary of a segmented object is more bent and more delicate when the compactness factor is 0.6 and more consistent with the characteristic of the ground object boundary, so that the parameters for performing multi-scale segmentation on the canopy height models of bamboos and eucalyptus in the forest area are obtained and are shown in the table 1. Optionally, the obtained segmented image may be further subjected to subsequent image processing, for example, a preset threshold is set, and a broken polygon of the crown boundary with the extraction error is extracted and merged into a large crown polygon of the same category.
TABLE 1
S302, obtaining a vector distribution map of the tree species in the forest region according to the segmentation image and the preset classification parameters.
Specifically, the computer device classifies the tree species in the forest region according to the obtained plurality of segmented images and preset classification parameters to obtain a classification map of the tree species in the forest region, and obtains a vector distribution map of the tree species in the forest region according to the classification map of the tree species in the forest region. Optionally, an image vectorization tool may be used to process the classification map of the tree species in the forest region, so as to obtain a vector distribution map of the tree species in the forest region.
In this embodiment, the computer device performs single-tree segmentation on the canopy height model by using a preset multi-scale segmentation method, so that the accuracy of the obtained multiple segmented images is improved, and the accuracy of the vector distribution map of the tree species in the forest region obtained according to the segmented images and preset classification parameters is improved.
Fig. 5 is a flowchart illustrating a tree species identification method according to another embodiment. FIG. 6 is a trend graph of the overall accuracy of classification of tree species in a forest area as a function of decision tree depth provided by one embodiment. FIG. 7 is a graph illustrating the trend of the overall accuracy of classification and the classification time of tree species in a forest area as a function of the number of decision trees, according to an embodiment. The embodiment relates to a specific implementation process for acquiring a vector distribution map of tree species in a forest region according to a segmented image and preset classification parameters. As shown in fig. 5, on the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S302 includes:
s501, determining the classification parameters according to a random forest classification method; the classification parameters include the depth of the decision tree and the number of the decision trees.
Specifically, the depths of different decision trees and the number of the decision trees have different influences on the classification precision of the tree species in the forest region, and the computer device can determine the depths of the decision trees and the number of the decision trees as classification parameters when the precision of the random forest model reaches the maximum value according to a random forest classification method. For example, when eucalyptus and bamboo in a forest region are classified, a trend graph of the classification overall accuracy along with the change of the depth of the decision tree is shown in fig. 6, a trend graph of the classification overall accuracy and the classification time along with the change of the number of the decision trees is shown in fig. 7, fig. 6 shows that the classification overall accuracy of eucalyptus and bamboo in the forest region is increased along with the increase of the depth of the tree, the accuracy is increased firstly and then is reduced, when the depth of the tree is increased to 30, the overall accuracy reaches a maximum value of 0.9, and it is stated that when the depth of the tree is 30, the classification overall accuracy of eucalyptus and bamboo in the forest region is effectively improved; fig. 7 shows that the overall accuracy of eucalyptus and bamboo classification in the forest area gradually increases with the change of the number of decision trees, when the number of decision trees exceeds 9, the classification time continuously increases, and when the number of decision trees reaches 30, the overall accuracy reaches the maximum, the classification time is relatively shortest, which indicates that the overall accuracy of eucalyptus and bamboo classification in the forest area is effectively improved when the number of decision trees is 30.
S502, determining a training sample set according to the segmented image; the training sample set includes images of various trees in the forest area.
Specifically, the computer device determines images including various trees in the forest region as a training sample set according to the obtained multiple segmented images of the forest region. For example, when the tree species to be recognized is bamboo and eucalyptus, the segmented images of bamboo, eucalyptus, masson pine, coriander, and oak in the forest area may be determined from the obtained plurality of segmented images of bamboo and eucalyptus in the forest area, and may be used as the training sample set.
S503, classifying the segmented images according to the training sample set and the classification parameters, and acquiring a vector distribution map of tree species in the forest region.
Specifically, the computer device classifies the obtained plurality of segmented images according to the training sample set and the classification parameters to obtain a grid classification map of tree species in the forest region, and performs vectorization processing on the obtained tree species classification map by using an image vectorization tool to obtain a vector distribution map of the tree species in the forest region. Optionally, the grid classification map of the tree species in the forest region may be a binary image, or may be a grayscale image or a color image. Optionally, the image vectorization tool may be a GIS grid image vectorization tool, or may be another image vectorization tool, as long as vectorization processing can be performed on the obtained tree type distribution map.
In this embodiment, the computer device determines a classification parameter according to a random forest method, determines a training sample set according to the obtained plurality of segmented images, classifies the obtained plurality of segmented images according to the determined training sample set and the classification parameter, and obtains a vector distribution map of tree species in a forest region, thereby improving the accuracy of the obtained vector distribution map of tree species in the forest region.
Fig. 8 is a flowchart illustrating a tree species identification method according to another embodiment. The embodiment relates to a specific implementation process for classifying segmented images according to a training sample set and classification parameters to obtain a vector distribution map of tree species in a forest region. As shown in fig. 8, on the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S503 includes:
s801, determining target characteristics according to a preset characteristic selection algorithm and data characteristics; the data features comprise point cloud features corresponding to the laser radar data and image features corresponding to the multispectral image.
Specifically, the computer device determines the target characteristics according to a preset characteristic selection algorithm, point cloud characteristics corresponding to the laser radar data and image characteristics corresponding to the multispectral image. Optionally, the preset feature selection algorithm may be a Boruta algorithm, and may also be another feature selection algorithm. Optionally, the point cloud features corresponding to the laser radar data include statistical features of normalized height features, intensity features, crown breadth and tree height, where the statistical features of the tree height include a maximum value, a minimum value, an average value, a standard deviation, and the like of the tree height; the image features corresponding to the multispectral image comprise spectral features, texture features and shape features, wherein the spectral features comprise pixel values of red, green, blue and near-infrared four wave bands, wave band mean values, normalized difference vegetation indexes and other various vegetation indexes, the texture features comprise mean values, standard deviations, homogeneity, contrast, non-similarity, entropy, angular second moment, correlation and the like of gray level co-occurrence matrixes, and the shape features comprise aspect ratios, lengths, widths, shape indexes, asymmetries, boundary lengths, densities and the like.
Optionally, as shown in fig. 9, on the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S801 may include:
and S901, sorting the data characteristics of the target tree species according to the importance of the data characteristics.
Specifically, the computer device randomizes the original attribute of each object in the data features according to a preset feature algorithm, adds a randomness component of the original attribute of each object, estimates the importance of the data features, and sorts the data features of the target trees according to the importance of the data features.
S902, respectively obtaining the first N data characteristics of each target tree species.
Specifically, the computer device obtains the top N data features of each target tree sort according to the obtained importance ranking of the data features of each target tree sort. Alternatively, N may be 20, 30 or determined according to actual conditions.
And S903, determining the union of the first N data characteristics of each target tree species as the target characteristics.
Specifically, the computer device obtains the first N data features of each target tree species, and determines a union of the first N data features of each target tree species as the target feature. For example, the target tree species in the forest area to be identified are bamboo and eucalyptus, the target features of the bamboo and eucalyptus are determined to be a union of the top 20 data features of the bamboo and eucalyptus, as shown in tables 2 and 3, table 2 is the target feature of the bamboo, and table 3 is the target feature of the eucalyptus.
TABLE 2
TABLE 3
In this embodiment, the computer device determines, according to the tree feature importance ranking of each target tree species, the same data feature among the top N data features of each target tree species as the target feature, thereby improving the accuracy of the determined target feature.
S802, classifying the segmented images according to the target features, the training sample set and the classification parameters, and obtaining a vector distribution map of tree species in the forest region.
Specifically, the computer device classifies each segmented image by adopting a random forest classification method according to the determined target feature, the training sample set and the classification parameters to obtain a grid classification map of tree species in the forest region, and vectorizes the obtained tree species classification map by adopting an image vectorization tool to obtain a vector distribution map of the tree species in the forest region.
In this embodiment, the computer device determines the target feature according to a preset feature selection algorithm and the data feature, removes some unimportant variables in the data feature, and improves the accuracy of the determined target feature, thereby improving the accuracy of the vector distribution area of the tree species in the forest region, which is obtained according to the target feature, the training sample set, and the classification parameter.
Fig. 10 is a flowchart illustrating a tree species identification method according to another embodiment. The embodiment relates to a specific implementation process for acquiring a canopy height model of a forest region according to laser radar data. As shown in fig. 10, on the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S202 includes:
and S1001, acquiring a digital elevation model and a digital surface model according to the laser radar data.
Specifically, the computer device resamples the discrete point cloud of the laser radar data by adopting a spatial interpolation method according to the acquired laser radar data to generate a digital elevation model and a digital surface model of the regular grid. Optionally, the sampling interval for resampling the discrete point cloud of lidar data may be 0.4 m.
S1002, acquiring the canopy height model according to the digital elevation model and the digital surface model.
Specifically, the computer device subtracts the digital elevation model and the digital surface model according to the acquired digital elevation model and the digital surface model to obtain a canopy height model. In this embodiment, the computer device obtains the digital elevation model and the digital surface model according to the laser radar data, and subtracts the digital elevation model and the digital surface model to obtain the canopy height model, so that the process of obtaining the digital elevation model and the digital surface model is very simple, and the calculation efficiency of obtaining the canopy height model is improved.
Fig. 11 is a flowchart illustrating a tree species identification method according to another embodiment. The embodiment relates to some processing of acquired laser radar data and multispectral data after acquiring laser radar data and multispectral images of a forest region. As shown in fig. 11, on the basis of the foregoing embodiment, as an optional implementation manner, after S201, the method further includes:
s1101, preprocessing the laser radar data according to a preset first processing method; the first processing method comprises a gross error rejection algorithm and/or a filtering classification algorithm.
Specifically, the computer device rejects elevation abnormal points in the laser radar data through a gross error rejection algorithm, and separates ground points and ground object points in the laser radar data through a filtering classification algorithm.
S1102, preprocessing the multispectral image by adopting a preset second processing method; the second processing method includes at least one of a correction method, a radiometric calibration method, an atmospheric correction method, and an image fusion method.
Specifically, the computer equipment adopts a correction method to perform high-precision orthorectification or real orthorectification on the acquired multispectral image to obtain a multispectral image which is unified with laser radar data in a space coordinate system; performing radiometric calibration on the acquired multispectral image by using a radiometric calibration method to obtain a radiance value; performing atmospheric correction on the multispectral image by adopting an atmospheric correction method; and obtaining the multispectral image with the same pixel size as the digital elevation model by adopting an image fusion method. Optionally, the adopted atmosphere correction method is to adopt a 6S model, a MODTRAN model or a FLAASH model.
In the embodiment, the computer device preprocesses the laser radar data according to a preset first processing method, so that the accuracy of the obtained canopy height model is improved; and preprocessing the multispectral image according to a preset second processing method, so that the precision of the tree species distribution map obtained according to the vector distribution map and the multispectral image is improved.
It should be understood that although the various steps in the flow charts of fig. 2-11 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-11 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.
Fig. 12 is a schematic structural diagram of a tree species identification apparatus according to an embodiment. As shown in fig. 12, the apparatus may include: a first obtaining module 10, a second obtaining module 11, a third obtaining module 12 and a determining module 13.
Specifically, the first obtaining module 10 obtains laser radar data and multispectral images of a forest area;
a second obtaining module 11, configured to obtain a canopy height model of the forest region according to the laser radar data;
a third obtaining module 12, configured to obtain a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
the determining module 13 is configured to obtain a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
Fig. 13 is a schematic structural diagram of a tree species identification apparatus according to an embodiment. On the basis of the embodiment shown in fig. 12, optionally, the third obtaining module 12 may include: a dividing unit 121 and a first acquiring unit 122.
Specifically, the segmentation unit 121 is configured to perform single-tree segmentation on the canopy height model by using a preset multi-scale segmentation method to obtain a plurality of segmented images;
the first obtaining unit 122 is configured to obtain a vector distribution map of tree species in the forest region according to the segmented image and the preset classification parameter.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
In an embodiment, the first obtaining unit 122 is specifically configured to determine the classification parameter according to a random forest classification method; the classification parameters comprise the depth of the decision tree and the number of the decision tree; determining a training sample set according to each segmented image; the training sample set comprises images of various trees in the forest region; and classifying the segmented images according to the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, the first obtaining unit 122 classifies each of the segmented images according to the training sample set and the classification parameters, and obtains a vector distribution map of tree species in the forest region, including: the first obtaining unit 122 determines a target feature according to a preset feature selection algorithm and a data feature; the data features comprise point cloud features corresponding to the laser radar data and image features corresponding to the multispectral image; and classifying the segmented images according to the target features, the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
On the basis of the above embodiment, the determining, by the first obtaining unit 122, the target feature according to the preset feature selection algorithm and the data feature includes: the first obtaining unit 122 sorts the data features of the target tree species according to the importance of the data features; respectively acquiring the first N data characteristics of each target tree species; and determining the same data feature in the first N data features of each target tree species as the target feature.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
Fig. 14 is a schematic structural diagram of a tree species identification device according to an embodiment. On the basis of the embodiment shown in fig. 12, optionally, the second obtaining module 11 may include: a second acquisition unit 111 and a third acquisition unit 112.
Specifically, the second obtaining unit 111 is configured to obtain a digital elevation model and a digital surface model according to the lidar data;
a third obtaining unit 112, configured to obtain the canopy height model according to the digital elevation model and the digital surface model.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
Fig. 15 is a schematic structural diagram of a tree species identification device according to an embodiment. On the basis of the above embodiment, as shown in fig. 15, the apparatus further includes: a first processing module 14 and a second processing module 15.
Specifically, the first processing module 14 is configured to perform preprocessing on the lidar data according to a preset first processing method; the first processing method comprises a gross error rejection algorithm and/or a filtering classification algorithm;
the second processing module 15 is configured to perform preprocessing on the multispectral image by using a preset second processing method; the second processing method includes at least one of a correction method, a radiometric calibration method, an atmospheric correction method, and an image fusion method.
The tree species identification apparatus provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
For the specific definition of the tree species identification device, reference may be made to the above definition of the tree species identification method, which is not described herein again. The modules in the tree identification device can be implemented in whole or in part by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring laser radar data and multispectral images of a forest area;
acquiring a canopy height model of the forest region according to the laser radar data;
acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
The implementation principle and technical effect of the computer device provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.
In one embodiment, a readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring laser radar data and multispectral images of a forest area;
acquiring a canopy height model of the forest region according to the laser radar data;
acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
The implementation principle and technical effect of the readable storage medium provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for identifying tree species, the method comprising:
acquiring laser radar data and multispectral images of a forest area;
acquiring a canopy height model of the forest region according to the laser radar data;
acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
2. The method according to claim 1, wherein the obtaining a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters comprises:
performing single-tree segmentation on the canopy height model by adopting a preset multi-scale segmentation method to obtain a segmented image;
and acquiring a vector distribution map of the tree species in the forest region according to the segmentation image and the preset classification parameters.
3. The method according to claim 2, wherein the obtaining a vector distribution map of tree species in the forest region according to the segmented image and the preset classification parameters comprises:
determining the classification parameters according to a random forest classification method; the classification parameters comprise the depth of the decision tree and the number of the decision tree;
determining a training sample set according to the segmented image; the training sample set comprises images of various trees in the forest region;
and classifying the segmented images according to the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region.
4. The method according to claim 3, wherein the classifying the segmented image according to the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region comprises:
determining target characteristics according to a preset characteristic selection algorithm and data characteristics; the data features comprise point cloud features corresponding to the laser radar data and image features corresponding to the multispectral image;
and classifying the segmented images according to the target features, the training sample set and the classification parameters to obtain a vector distribution map of tree species in the forest region.
5. The method of claim 4, wherein determining the target feature based on a pre-set feature selection algorithm and the data feature comprises:
sorting the data characteristics of the target tree species according to the importance of the data characteristics;
respectively acquiring the first N data characteristics of each target tree species;
and determining the union of the first N data characteristics of each target tree species as the target characteristics.
6. The method according to any one of claims 1-5, wherein said obtaining a canopy height model of the forest area from the lidar data comprises:
acquiring a digital elevation model and a digital surface model according to the laser radar data;
and acquiring the canopy height model according to the digital elevation model and the digital surface model.
7. The method according to any one of claims 1-5, wherein after acquiring the lidar data and the multi-spectral image of the forest area, the method further comprises:
preprocessing the laser radar data according to a preset first processing method; the first processing method comprises a gross error rejection algorithm and/or a filtering classification algorithm;
preprocessing the multispectral image by adopting a preset second processing method; the second processing method includes at least one of a correction method, a radiometric calibration method, an atmospheric correction method, and an image fusion method.
8. An apparatus for identifying tree species, the apparatus comprising:
the first acquisition module is used for acquiring laser radar data and multispectral images of a forest area;
the second acquisition module is used for acquiring a canopy height model of the forest region according to the laser radar data;
the third acquisition module is used for acquiring a vector distribution map of tree species in the forest region according to the canopy height model and preset classification parameters;
the determining module is used for obtaining a tree species distribution map according to the vector distribution map and the multispectral image; the tree species distribution map is used for identifying a target tree species in the forest region.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method according to any of claims 1-7.
10. A readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811243826.5A CN111091030A (en) | 2018-10-24 | 2018-10-24 | Tree species identification method and device, computer equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811243826.5A CN111091030A (en) | 2018-10-24 | 2018-10-24 | Tree species identification method and device, computer equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111091030A true CN111091030A (en) | 2020-05-01 |
Family
ID=70391614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811243826.5A Pending CN111091030A (en) | 2018-10-24 | 2018-10-24 | Tree species identification method and device, computer equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111091030A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112183489A (en) * | 2020-11-04 | 2021-01-05 | 长光禹辰信息技术与装备(青岛)有限公司 | Pine color-changing standing tree identification and positioning method, device, equipment and storage medium |
| CN112577907A (en) * | 2020-11-18 | 2021-03-30 | 上海市园林科学规划研究院 | Urban green land tree crown loss rate calculation method |
| CN112819066A (en) * | 2021-01-28 | 2021-05-18 | 北京林业大学 | Res-UNet single tree species classification technology |
| CN114005032A (en) * | 2021-10-28 | 2022-02-01 | 广州市城市规划勘测设计研究院 | Urban street tree single tree parameter extraction method and device and terminal equipment |
| CN116704333A (en) * | 2023-05-19 | 2023-09-05 | 中国电建集团江西省电力设计院有限公司 | Single tree detection method based on laser point cloud data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354534A (en) * | 2015-09-29 | 2016-02-24 | 南京林业大学 | Tree species classification method based on multi-source simultaneous high-resolution remote sensing data |
| CN105590313A (en) * | 2015-11-12 | 2016-05-18 | 北京林业大学 | Forest three- dimensional canopy morphological structure extraction method on the basis of active contour model |
| CN107085710A (en) * | 2017-04-26 | 2017-08-22 | 长江空间信息技术工程有限公司(武汉) | A kind of single wooden extraction method based on multispectral LiDAR data |
| CN107727598A (en) * | 2017-10-13 | 2018-02-23 | 中国科学院上海技术物理研究所 | A kind of transmitted spectrum imaging method for aqueous hyaline tissue |
-
2018
- 2018-10-24 CN CN201811243826.5A patent/CN111091030A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354534A (en) * | 2015-09-29 | 2016-02-24 | 南京林业大学 | Tree species classification method based on multi-source simultaneous high-resolution remote sensing data |
| CN105590313A (en) * | 2015-11-12 | 2016-05-18 | 北京林业大学 | Forest three- dimensional canopy morphological structure extraction method on the basis of active contour model |
| CN107085710A (en) * | 2017-04-26 | 2017-08-22 | 长江空间信息技术工程有限公司(武汉) | A kind of single wooden extraction method based on multispectral LiDAR data |
| CN107727598A (en) * | 2017-10-13 | 2018-02-23 | 中国科学院上海技术物理研究所 | A kind of transmitted spectrum imaging method for aqueous hyaline tissue |
Non-Patent Citations (5)
| Title |
|---|
| 崔小芳;刘正军;: "基于随机森林分类方法和多源遥感数据的湿地植被精细分类", no. 08, pages 123 - 126 * |
| 王志慧等: "基于scikit-learn的机器学习 算法与实践", vol. 1, 黄河水利出版社, pages: 118 - 190 * |
| 陶江;刘丽娟;丁友丽;王雪;彭琼;肖文惠;: "高光谱和LiDAR数据融合在树种识别上的应用", no. 08, pages 216 - 218 * |
| 陶江;刘丽娟;庞勇;李登秋;冯云云;王雪;丁友丽;彭琼;肖文惠;: "基于机载激光雷达和高光谱数据的树种识别方法", vol. 1, no. 02, pages 129 - 138 * |
| 鲁续坤: "基于机载LiDAR和高光谱数据的树种分类及三维显示", no. 9, pages 049 - 2 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112183489A (en) * | 2020-11-04 | 2021-01-05 | 长光禹辰信息技术与装备(青岛)有限公司 | Pine color-changing standing tree identification and positioning method, device, equipment and storage medium |
| CN112183489B (en) * | 2020-11-04 | 2023-03-21 | 长光禹辰信息技术与装备(青岛)有限公司 | Pine color-changing standing tree identification and positioning method, device, equipment and storage medium |
| CN112577907A (en) * | 2020-11-18 | 2021-03-30 | 上海市园林科学规划研究院 | Urban green land tree crown loss rate calculation method |
| CN112819066A (en) * | 2021-01-28 | 2021-05-18 | 北京林业大学 | Res-UNet single tree species classification technology |
| CN114005032A (en) * | 2021-10-28 | 2022-02-01 | 广州市城市规划勘测设计研究院 | Urban street tree single tree parameter extraction method and device and terminal equipment |
| CN114005032B (en) * | 2021-10-28 | 2022-06-14 | 广州市城市规划勘测设计研究院 | Method and device for extracting single tree parameters of urban street tree and terminal equipment |
| CN116704333A (en) * | 2023-05-19 | 2023-09-05 | 中国电建集团江西省电力设计院有限公司 | Single tree detection method based on laser point cloud data |
| CN116704333B (en) * | 2023-05-19 | 2024-03-12 | 中国电建集团江西省电力设计院有限公司 | Single tree detection method based on laser point cloud data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111091030A (en) | Tree species identification method and device, computer equipment and readable storage medium | |
| CN110852225B (en) | Remote sensing image mangrove forest extraction method and system based on deep convolutional neural network | |
| US9064150B2 (en) | Aerial image segmentation for refineries | |
| US20180081035A1 (en) | Method and device for filtering point cloud data | |
| CN112884764B (en) | Method and device for extracting land block in image, electronic equipment and storage medium | |
| KR100935857B1 (en) | System for generating 3d forest geographical information data using aerial lidar data and digital aerial photo data and method therefor | |
| Röder et al. | Application of optical unmanned aerial vehicle-based imagery for the inventory of natural regeneration and standing deadwood in post-disturbed spruce forests | |
| CN110728197A (en) | Single-tree-level tree species identification method based on deep learning | |
| CN115909113B (en) | Method for investigating forestry harmful organisms through unmanned aerial vehicle remote sensing monitoring | |
| CN105590313A (en) | Forest three- dimensional canopy morphological structure extraction method on the basis of active contour model | |
| CN114240980A (en) | Unmanned aerial vehicle image sub-field boundary extraction method and device | |
| CN109657540B (en) | Withered tree positioning method and system | |
| Amiri et al. | Feature relevance assessment of multispectral airborne LiDAR data for tree species classification | |
| JP5352435B2 (en) | Classification image creation device | |
| CN116994029A (en) | Fusion classification method and system for multi-source data | |
| Xuegang et al. | Object-oriented automatic identification of forest gaps using digital orthophoto maps and LiDAR data | |
| CN115661471A (en) | Method and device for extracting laver culture area, readable storage medium and electronic equipment | |
| CN114463642A (en) | Cultivated land plot extraction method based on deep learning | |
| Tompalski et al. | A comparison of LiDAR and image-derived canopy height models for individual tree crown segmentation with object based image analysis | |
| Marin et al. | Individual Olive Tree Detection in RGB Images | |
| Aleshko et al. | Development of methodology for automated determination of forest parameters based on data from unmanned aerial vehicles | |
| Erdem et al. | Reproducible Extraction of Building Footprints from Airborne LiDAR Data: A Demo Paper | |
| Fan et al. | An improved Deeplab based model for extracting cultivated land information from high definition remote sensing images | |
| CN111429509A (en) | Centralized measuring and calculating method for height of target object | |
| CN112396615A (en) | Individual stumpage segmentation method and device based on adjustable hemispherical canopy camera |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200501 |
|
| RJ01 | Rejection of invention patent application after publication |