CN113011266A - Sky-ground integrated pine wood nematode disease epidemic situation remote sensing monitoring method - Google Patents
Sky-ground integrated pine wood nematode disease epidemic situation remote sensing monitoring method Download PDFInfo
- Publication number
- CN113011266A CN113011266A CN202110197983.2A CN202110197983A CN113011266A CN 113011266 A CN113011266 A CN 113011266A CN 202110197983 A CN202110197983 A CN 202110197983A CN 113011266 A CN113011266 A CN 113011266A
- Authority
- CN
- China
- Prior art keywords
- wood nematode
- pine wood
- nematode disease
- sample
- remote sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 201000010099 disease Diseases 0.000 title claims abstract description 191
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 191
- 241000243771 Bursaphelenchus xylophilus Species 0.000 title claims abstract description 149
- 238000012544 monitoring process Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 49
- 238000011835 investigation Methods 0.000 claims abstract description 22
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 64
- 235000008331 Pinus X rigitaeda Nutrition 0.000 claims description 49
- 235000011613 Pinus brutia Nutrition 0.000 claims description 49
- 241000018646 Pinus brutia Species 0.000 claims description 49
- 230000003595 spectral effect Effects 0.000 claims description 29
- 238000012795 verification Methods 0.000 claims description 18
- 238000001228 spectrum Methods 0.000 claims description 15
- 238000003066 decision tree Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000005855 radiation Effects 0.000 claims description 3
- 238000002310 reflectometry Methods 0.000 claims description 3
- 241000244206 Nematoda Species 0.000 claims 1
- 238000012163 sequencing technique Methods 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 2
- 238000009472 formulation Methods 0.000 abstract description 2
- 239000000203 mixture Substances 0.000 abstract description 2
- 230000008030 elimination Effects 0.000 description 4
- 238000003379 elimination reaction Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 1
- 206010035148 Plague Diseases 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000010354 integration 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to a remote sensing monitoring method for the pine wood nematode disease in the sky and the ground, which obtains the original data of multi-source monitoring aiming at the pine wood nematode disease by adopting a method of integrating a satellite remote sensing image, a satellite remote sensing image and ground investigation, and the remote sensing interpretation sample aiming at the pine wood nematode disease epidemic situation area is obtained by preprocessing the obtained pine wood nematode disease epidemic situation area data set, and then the pine wood nematode disease epidemic situation monitoring result aiming at the pine wood nematode disease epidemic situation area is obtained by processing according to the obtained pine wood nematode disease epidemic situation area data set and the remote sensing interpretation sample, therefore, the defect of a single monitoring data source is made up, the comprehensive monitoring of the pine wood nematode disease epidemic situation area is realized, and the monitoring accuracy and the monitoring efficiency are improved, so that the tree withered and dead due to the pine wood nematode disease can be found more comprehensively, quickly and accurately, and detailed data guarantee is provided for the subsequent formulation of a control scheme.
Description
Technical Field
The invention relates to the field of remote sensing monitoring, in particular to a remote sensing monitoring method for a sky-ground integrated pine wood nematode disease situation.
Background
Pine wilt disease is also known as pine wilt disease and is one of the most harmful diseases in forest ecosystems. The disease has the characteristics of wide transmission path, high speed, great prevention and treatment difficulty and the like, and is called cancer of pine trees. The pine wood nematode disease not only causes serious economic loss, but also harms the development of forest resources and the stability of ecological environment. The monitoring and investigation of the pine wood nematode disease is the work basis of the prevention and control of the pine wood nematode disease, and the accurate and efficient monitoring of the pine wood nematode disease has important significance for comprehensively mastering the disease information and inhibiting the spread of the pine wood nematode disease in time.
The existing pine wood nematode disease monitoring method mainly adopts manual investigation, although a large amount of data are accumulated for pine wood nematode disease management, the method consumes long time and is complicated in procedure, and the occurrence and development conditions of the disease cannot be accurately mastered.
Certainly, there is also pine wood nematode disease epidemic situation monitoring method based on remote sensing technology at present, and the distribution and the location of the tree dying in pine wood nematode disease area can be monitored rapidly and timely by analyzing and processing the optical satellite remote sensing image obtained by the remote sensing technology, so that the monitoring cost is reduced while the monitoring efficiency is improved. However, because the optical satellite remote sensing image is greatly influenced by weather and is limited by spatial resolution or spectral resolution, it is difficult to obtain ideal data meeting the requirement of accurate monitoring of pine wood nematode plague.
Therefore, how to realize the comprehensive, rapid and accurate monitoring of the pine wood nematode disease epidemic situation and improve the monitoring efficiency becomes a technical problem which needs to be solved urgently in the current pine wood nematode disease epidemic situation monitoring.
Disclosure of Invention
The invention aims to solve the technical problem of providing a remote sensing monitoring method for the aerial integrated bursaphelenchus xylophilus disease epidemic situation in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows: a remote sensing monitoring method for the planting situation of the pine wood nematode disease integrated in the sky is characterized by comprising the following steps of S1-S4:
step S1, respectively acquiring satellite remote sensing image original data of a target area to be monitored, unmanned aerial vehicle hyperspectral image original data of a pine wood nematode disease epidemic situation area and ground investigation original data of the pine wood nematode disease epidemic situation area, and forming a pine wood nematode disease epidemic situation area original data set by the acquired satellite remote sensing image original data, unmanned aerial vehicle hyperspectral image original data and the ground investigation original data; the target area to be monitored comprises a pine wood nematode disease epidemic situation area, an acquisition area corresponding to ground inspection original data is consistent with an acquisition area corresponding to unmanned aerial vehicle hyperspectral image original data, and the ground inspection original data at least comprises a position of a tree which is killed by the pine wood nematode disease in the pine wood nematode disease epidemic situation area and a picture of the tree which is killed by the pine wood nematode disease;
step S2, respectively preprocessing the original data of the satellite remote sensing image, the original data of the hyperspectral image of the unmanned aerial vehicle and the original data of the ground investigation in the original data set of the pine wood nematode disease area, correspondingly obtaining the original data of the satellite remote sensing image, the hyperspectral image of the unmanned aerial vehicle and the ground investigation data, and forming a pine wood nematode disease area data set by using the obtained data;
step S3, establishing a remote sensing interpretation sample aiming at the pine wood nematode disease epidemic situation area; the remote sensing interpretation sample comprises an image sample, a spectrum sample and sample point attribute information aiming at each type of ground object in the pine wood nematode disease situation area to be monitored;
and S4, processing the pine wood nematode disease situation monitoring result aiming at the pine wood nematode disease situation area according to the obtained pine wood nematode disease situation area data set and the remote sensing interpretation sample.
In the method for remotely sensing and monitoring the pine wood nematode disease in the sky and the ground, in step S2, the preprocessing of the raw data of the hyperspectral image of the unmanned aerial vehicle includes the following steps:
carrying out format conversion processing on the unmanned aerial vehicle hyperspectral image original data in the pine wood nematode disease situation area original data set; the format conversion processing comprises the steps of converting cube data of the hyperspectral image original data of the unmanned aerial vehicle into ENVI format data and converting the ENVI format data into JPG format data;
carrying out mosaic splicing treatment on the image of the original data of the hyperspectral image of the unmanned aerial vehicle after format conversion; the mosaic splicing processing comprises image alignment processing, feature point generation processing, point cloud generation processing, texture calculation processing, grid construction processing and mosaic model generation processing of the converted JPG format data.
Further, in the remote sensing monitoring method for the skies-ground integrated pine wood nematode disease situation, in step S2, the preprocessing of the original data of the satellite remote sensing image includes geometric correction processing, radiation correction processing and atmospheric correction processing for the original data of the satellite remote sensing image.
Still further, in the remote sensing monitoring method for the skimming integrated pine wood nematode disease, in step S2, the preprocessing process of the ground-based inspection original data includes:
inputting ground investigation original data aiming at the tree withered and dead due to the pine wilt disease in the pine wilt disease epidemic situation area;
and, carry on the standardized processing to the primitive data of this ground investigation entered; the ground-stepping raw data comprises a ground-stepping photo file path, a sample category, equipment parameters, weather parameters and remark information.
Further improved, in the remote sensing monitoring method for the skies and lands integrated bursaphelenchus xylophilus disease epidemic situation, the establishment process of the remote sensing interpretation sample comprises the following steps of S31-S35:
step S31, selecting a surface feature sample set aiming at the pine wood nematode disease situation area to be monitored, and acquiring the position information and surface feature sample photos of each surface feature sample in the surface feature sample set in a surface feature examination mode; the surface feature samples in the surface feature sample set comprise healthy pine samples, dead tree samples caused by the pine wilt disease, broad-leaved tree samples, bare surface samples, building roof samples and road samples in a pine wilt disease epidemic situation area to be monitored;
step S32, on the basis of the preprocessed unmanned aerial vehicle hyperspectral image data, taking a pixel of the spatial position information of the ground stepping position point on the unmanned aerial vehicle hyperspectral image data as a center, and intercepting a rectangular range of 3 x 3 pixels as an image sample for each type of ground object in the pine wood nematode disease epidemic situation area to be monitored; wherein, the rectangular range of the 3 x 3 image elements comprises 9 image elements;
step S33, taking the intercepted average spectral characteristic curve of 9 pixels in the rectangular range of the 3 x 3 pixels as a spectral sample for each category of ground objects in the pine wood nematode disease situation area to be monitored;
and step S34, establishing a space data set of the remote sensing interpretation sample by using the space position information of the ground treading point position, endowing each ground feature sample with a unique sample code, and forming the remote sensing interpretation sample by the space position information of each ground treading point position, the image sample of each type of ground feature and the spectrum sample of each type of ground feature.
Further, in the remote sensing monitoring method for the pine wood nematode disease in the sky-ground integration manner, in step S4, a monitoring result of the pine wood nematode disease in the pine wood nematode disease area is obtained by processing the following steps S41 to S411:
step S41, superposing the pine forest class vector data in the forest resource second-class survey database with the multi-source remote sensing image by adopting a GIS superposition analysis method, and intercepting the image of the pine forest class in the pine wood nematode disease epidemic situation area to be monitored; the multisource remote sensing image is satellite remote sensing image data and unmanned aerial vehicle hyperspectral image data in the pine wood nematode disease situation area data set;
step S42, quantitative inversion extraction is carried out on hyperspectral image data of the unmanned aerial vehicle by utilizing the spectrum characteristic index model of the pine wilt disease dead tree, and a hyperspectral-based pine wilt disease dead tree extraction result is obtained;
step S43, taking the obtained pine wood nematode disease dead tree extraction result based on the hyperspectrum and other surface feature samples of the non-pine wood nematode disease dead tree sample in the remote sensing interpretation sample as area classification samples of the pine wood nematode disease epidemic situation area satellite remote sensing image data;
step S44, randomly classifying the region classification sample into two sub-region samples, taking one of the sub-region samples as a training sample group, and taking the other sub-region sample as a verification sample group;
step S45, respectively extracting spectral characteristic index sets and texture characteristic index sets of various ground features in the training sample group aiming at the satellite remote sensing image data preprocessed in the step S2, and forming training sample files by all the extracted characteristic index sets; the training sample file is marked as D, the total number of the ground feature type types in the training sample file D is v, and v is more than 1;
step S46, corresponding to any characteristic index set in the training sample fileThe characteristic indexes are used as category classification attributes, and all characteristic index values in the training sample file are sequenced to obtain a sequenced characteristic index value sequence; n characteristic index sets are set in a training sample file D, and the characteristic index corresponding to any one characteristic index set is marked as An，1≤n≤N；
Step S47, respectively calculating the information gain of each characteristic index in the training sample file; wherein:
wherein p isiRepresenting the probability of the ith type ground object sample in the training sample file D; infr (D) represents the information entropy of the training sample file D;
InforA(D) representing characteristic index A in training sample file DnEntropy of the information contained, DjRepresenting characteristic index A in training sample file DnA subsample set formed when the value is j; InforGain (A)n) InforGain (A) represents characteristic index A in training sample file DnThe corresponding information gain;
step S48, calculating the splitting information degree of each characteristic index in the training sample file, and calculating the information gain rate of the characteristic index by the information gain of each characteristic index and the corresponding splitting information degree; wherein:
wherein,
indicates a characteristic index AnIs split information degree, gain ratio (A)n) GainRatio (A) represents a characteristic index AnThe information gain rate of (d);
step S49, selecting the characteristic index with the maximum information gain rate as the split node attribute from all the characteristic indexes in the training sample file, and dividing the training sample file into two sample file subsets by using the split node attribute;
s410, repeatedly processing each sample file subset according to the mode of S46-S49 until the ground feature types of the samples in each sample file subset are all single types, and taking the obtained training result as a decision tree classification rule set;
and S411, according to a decision tree method, classifying and extracting the satellite remote sensing image data of the pine wood nematode disease epidemic situation area by using the obtained classification rule set, and taking the pine wood nematode disease dying tree pattern spots in the classification extraction result as the pine wood nematode disease epidemic situation monitoring result aiming at the pine wood nematode disease epidemic situation area.
Further, in the remote sensing monitoring method for the skies and the lands integrated pine wood nematode disease epidemic situation, in the step S42, the construction process of the spectrum characteristic index model of the dead tree of the pine wood nematode disease comprises the following steps b 1-b 4:
b1, respectively performing envelope elimination on the spectrum curves of all the surface features in the surface feature sample set by adopting an envelope elimination method to obtain enhanced spectrum characteristic curves of all the surface features;
b2, extracting the difference wave bands of the enhanced spectral characteristic curve of the tree dead due to the pine wilt disease in the surface feature sample set and the enhanced spectral characteristic curves of other surface features in a manual interpretation mode; wherein, the other ground features are the ground features except for the dead tree caused by the pine wilt disease in the ground feature sample set;
b3, taking the extracted difference wave band as a characteristic wave band, and taking the characteristic wave band as spectral characteristic analysis information of the tree withered due to the pine wood nematode disease;
b4, establishing a linear model by using the characteristic wave band, and taking the linear model as a spectral characteristic index model of the tree withered and dead by the pine wood nematode disease; wherein, the spectral characteristic index of the dead tree caused by the pine wilt disease is marked as IBx：
Wherein R is526、R534And R654The spectral reflectivities at 526nm, 534nm and 654nm are hyperspectral data for the unmanned aerial vehicle.
Further improved, in the remote sensing monitoring method for the skies and lands integrated pine wood nematode disease, the step S411 further includes steps a 1-a 3:
step a1, classifying and extracting the verification sample group by using the decision tree classification rule set to obtain a sample quality verification data result of the pine wood nematode disease dead tree;
step a2, judging according to the obtained sample quality verification data result:
when the obtained sample verification data result meets the preset precision requirement, turning to the step a 3; otherwise, correcting the training sample set, and executing the step S45-step S410 until the re-extracted sample verification data result meets the preset precision requirement, and executing the step a 3;
in step a3, the current decision tree classification rule set is used as the classification rule set in step S411.
Preferably, in the remote sensing monitoring method for the skies and the lands integrated pine wood nematode disease, the original hyperspectral image data of the unmanned aerial vehicle is acquired by a Roc unmanned aerial vehicle with the model of CW-10, and the original satellite remote sensing image data is satellite remote sensing image data with the spatial resolution of 1 m.
Compared with the prior art, the invention has the advantages that: the invention obtains multisource monitoring original data aiming at the pine wood nematode disease epidemic situation area by integrating the satellite remote sensing image technology, the unmanned aerial vehicle hyperspectral image technology and the ground investigation method, obtains the remote sensing interpretation sample aiming at the pine wood nematode disease epidemic situation area by preprocessing the obtained pine wood nematode disease epidemic situation area data set, and then obtains the pine wood nematode disease epidemic situation monitoring result aiming at the pine wood nematode disease epidemic situation area by processing according to the obtained pine wood nematode disease area data set and the remote sensing interpretation sample, thereby not only making up the defect of a single monitoring data source, for example, avoiding the condition that a single optical satellite remote sensing image is easily influenced by weather, realizing the comprehensive monitoring of the pine wood nematode disease epidemic situation area, but also improving the monitoring accuracy and monitoring efficiency of the pine wood nematode disease epidemic situation, thereby being convenient to more comprehensively, quickly and accurately discover the pine wood nematode disease death tree, and detailed data guarantee is provided for the subsequent formulation of a prevention and treatment scheme.
Drawings
Fig. 1 is a schematic flow chart of a remote sensing monitoring method for the skies-ground integrated pine wood nematode disease in the embodiment of the invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying examples.
The embodiment provides a remote sensing monitoring method for pine wood nematode disease, in particular to a remote sensing monitoring method for pine wood nematode disease integrated in a sky. Specifically, as shown in fig. 1, the remote sensing monitoring method for the skies-ground integrated pine wood nematode disease situation in the embodiment includes the following steps S1-S4:
step S1, respectively acquiring satellite remote sensing image original data of a target area to be monitored, unmanned aerial vehicle hyperspectral image original data of a pine wood nematode disease epidemic situation area and ground investigation original data of the pine wood nematode disease epidemic situation area, and forming a pine wood nematode disease epidemic situation area original data set by the acquired satellite remote sensing image original data, unmanned aerial vehicle hyperspectral image original data and the ground investigation original data; the hyperspectral image original data of the unmanned aerial vehicle is acquired by a Roc unmanned aerial vehicle with the model of CW-10, the satellite remote sensing image original data is satellite remote sensing image data with the spatial resolution of 1m, a target area to be monitored comprises a pine wilt disease epidemic situation area, an acquisition area corresponding to ground treading original data is consistent with an acquisition area corresponding to the hyperspectral image original data of the unmanned aerial vehicle, and the ground treading original data at least comprises the position of a tree dying due to the pine wilt disease in the pine wilt disease epidemic situation area and a picture of the tree dying due to the pine wilt disease; the positions of the trees withered due to the pine wilt disease correspond to the photos of the trees withered due to the pine wilt disease one by one;
step S2, respectively preprocessing the original data of the satellite remote sensing image, the original data of the hyperspectral image of the unmanned aerial vehicle and the original data of the ground investigation in the original data set of the pine wood nematode disease epidemic situation area, correspondingly obtaining the original data of the satellite remote sensing image, the hyperspectral image of the unmanned aerial vehicle and the ground investigation data, and forming an epidemic situation data set of the pine wood nematode disease area by using the obtained data; in step S2, the preprocessing processes for the acquired satellite remote sensing image raw data, the unmanned aerial vehicle hyperspectral image raw data, and the ground scout raw data are as follows:
(1) the preprocessing process of the satellite remote sensing image original data comprises geometric correction processing, radiation correction processing and atmospheric correction processing aiming at the satellite remote sensing image original data.
(2) The preprocessing process of the hyperspectral image original data of the unmanned aerial vehicle is as follows:
carrying out format conversion processing on the unmanned aerial vehicle hyperspectral image original data in the pine wood nematode disease situation area original data set; the format conversion processing comprises the steps of converting cube data of the hyperspectral image original data of the unmanned aerial vehicle into ENVI format data and converting the ENVI format data into JPG format data;
carrying out mosaic splicing treatment on the image of the hyperspectral image data of the unmanned aerial vehicle after format conversion; the mosaic splicing processing comprises image alignment processing, feature point generation processing, point cloud generation processing, texture calculation processing, grid construction processing and mosaic model generation processing of the converted JPG format data.
(3) The preprocessing process of the ground-stepping raw data is as follows:
inputting ground investigation original data aiming at the tree withered and dead due to the pine wilt disease in the pine wilt disease epidemic situation area; and, carry on the standardized processing to the primitive data of this ground investigation entered; the ground-stepping raw data comprises a ground-stepping photo file path, a sample category, equipment parameters, weather parameters and remark information.
Step S3, establishing a remote sensing interpretation sample aiming at the pine wood nematode disease epidemic situation area; the remote sensing interpretation sample comprises an image sample, a spectrum sample and sample point attribute information aiming at each type of ground object in the pine wood nematode disease situation area to be monitored; specifically, in this embodiment, the process of establishing the remote sensing interpretation sample includes the following steps S31 to S35:
step S31, selecting a surface feature sample set aiming at the pine wood nematode disease situation area to be monitored, and acquiring the position information and surface feature sample photos of each surface feature sample in the surface feature sample set in a surface feature examination mode; the surface feature samples in the surface feature sample set comprise healthy pine samples, dead tree samples caused by the pine wilt disease, broad-leaved tree samples, bare surface samples, building roof samples and road samples in a pine wilt disease epidemic situation area to be monitored;
step S32, on the basis of the preprocessed unmanned aerial vehicle hyperspectral image data, taking a pixel of the spatial position information of the ground stepping position point on the unmanned aerial vehicle hyperspectral image data as a center, and intercepting a rectangular range of 3 x 3 pixels as an image sample for each type of ground object in the pine wood nematode disease epidemic situation area to be monitored; wherein, the rectangular range of the 3 x 3 image elements comprises 9 image elements;
step S33, taking the intercepted average spectral characteristic curve of 9 pixels in the rectangular range of the 3 x 3 pixels as a spectral sample for each category of ground objects in the pine wood nematode disease situation area to be monitored;
and step S34, establishing a space data set of the remote sensing interpretation sample by using the space position information of the ground treading point position, endowing each ground feature sample with a unique sample code, and forming the remote sensing interpretation sample by the space position information of each ground treading point position, the image sample of each type of ground feature and the spectrum sample of each type of ground feature.
And S4, processing the pine wood nematode disease situation monitoring result aiming at the pine wood nematode disease situation area according to the obtained pine wood nematode disease area situation data set and the remote sensing interpretation sample. The pine wood nematode disease monitoring result aiming at the pine wood nematode disease area is obtained by processing according to the following steps S41-S411:
step S41, superposing the pine forest class vector data in the forest resource second-class survey database with the multi-source remote sensing image by adopting a GIS superposition analysis method, and intercepting the image of the pine forest class in the pine wood nematode disease epidemic situation area to be monitored; the multisource remote sensing image is satellite remote sensing image data and unmanned aerial vehicle hyperspectral image data in the pine wood nematode disease situation area data set; for example, in the embodiment, the geographic information system data processing software ArcGIS is used for carrying out spatial position superposition under a unified spatial coordinate system and carrying out data cutting, wherein the cutting result is the captured image of the pine forest class in the pine wood nematode disease epidemic situation area to be monitored;
step S42, quantitative inversion extraction is carried out on hyperspectral image data of the unmanned aerial vehicle by utilizing the spectrum characteristic index model of the pine wilt disease dead tree, and a hyperspectral-based pine wilt disease dead tree extraction result is obtained; the construction process of the spectral characteristic index model for the tree withered and dead due to the pine wilt disease in the training sample group comprises the following steps b 1-b 4:
b1, respectively performing envelope elimination on the spectrum curves of all the surface features in the surface feature sample set by adopting an envelope elimination method to obtain enhanced spectrum characteristic curves of all the surface features;
b2, extracting the difference wave bands of the enhanced spectral characteristic curve of the tree dead due to the pine wilt disease in the surface feature sample set and the enhanced spectral characteristic curves of other surface features in a manual interpretation mode; wherein, the other ground features are the ground features except for the dead tree caused by the pine wilt disease in the ground feature sample set;
b3, taking the extracted difference wave band as a characteristic wave band, and taking the characteristic wave band as spectral characteristic analysis information of the tree withered due to the pine wood nematode disease;
b4, establishing a linear model by using the characteristic wave band, and taking the linear model as a spectral characteristic index model of the tree withered and dead by the pine wood nematode disease; wherein, the spectral characteristic index of the dead tree caused by the pine wilt disease is marked as IBx：
Wherein R is526、R534And R654Spectral reflectivities at 526nm, 534nm and 654nm for hyperspectral data of the unmanned aerial vehicle;
step S43, taking the obtained pine wood nematode disease dead tree extraction result based on the hyperspectrum and other surface feature samples of the non-pine wood nematode disease dead tree sample in the remote sensing interpretation sample as area classification samples of the pine wood nematode disease epidemic situation area satellite remote sensing image data;
step S44, randomly classifying the region classification sample into two sub-region samples, taking one of the sub-region samples as a training sample group, and taking the other sub-region sample as a verification sample group;
step S45, respectively extracting spectral characteristic index sets and texture characteristic index sets of various ground features in the training sample group aiming at the satellite remote sensing image data preprocessed in the step S2, and forming training sample files by all the extracted characteristic index sets; the training sample file is marked as D, the total number of the surface feature types in the training sample file D is v, v is more than 1, namely, the training sample file is provided with a v-type surface feature type sample which comprises a tree withered and dead due to the pine wilt disease;
step S46, using the feature index corresponding to any feature index set in the training sample file D as the classification attribute of the category, and using all the feature index values in the training sample fileSequencing to obtain a sequenced characteristic index value sequence; the sorting of all the characteristic index values in the training sample file can be performed in an ascending sorting mode or a descending sorting mode according to the numerical value of the characteristic index values according to needs; setting N characteristic index sets in a training sample file D, wherein the characteristic index mark corresponding to any one characteristic index set is An，1≤n≤N；
Step S47, respectively calculating the information gain of each characteristic index in the training sample file; wherein:
wherein p isiRepresenting the probability of the ith type ground object sample in the training sample file D; infr (D) represents the information entropy of the training sample file D;
InforA(D) representing characteristic index A in training sample file DnEntropy of the information contained, DjRepresenting characteristic index A in training sample file DnA subsample set formed when the value is j; InforGain (A)n) Representing characteristic index A in training sample file DnThe corresponding information gain;
step S48, respectively calculating the splitting information degree of each characteristic index in the training sample file D, and obtaining the information gain rate of the characteristic index through the common calculation of the information gain of each characteristic index and the corresponding splitting information degree; wherein:
wherein,
indicates a characteristic index AnIs split information degree, gain ratio (A)n) Indicates a characteristic index AnThe information gain rate of (d);
step S49, selecting the characteristic index with the maximum information gain rate as the split node attribute from all the characteristic indexes in the training sample file D, and dividing the training sample file D into two sample file subsets by using the split node attribute; for example, in the training sample file D, all the characteristic indexes A are obtained1～ANIn (1), assume its characteristic index A6Maximum information gain ratio (A)6) Then, the characteristic index A is used here6As a split node attribute, and dividing the training sample file D into two sample file subsets by using the split node attribute, specifically dividing the training sample file D into the sample file subset D1And sample file subset D2；
S410, repeatedly processing each sample file subset according to the mode of S46-S49 until the ground feature types of the samples in each sample file subset are all single types, and taking the obtained training result as a decision tree classification rule set; i.e. for the sample file subset D1Sequentially processed in the manner of steps S46-S49, and aiming at the sample file subset D2Processing the sample files sequentially according to the steps S46 to S49 until D in the sample file subset1The ground object class of the sample is a single class and D in the sample file subset2When the ground feature types of the samples are all single types, the obtained training result is used as a decision tree classification rule set;
step S411, according to a conventional decision tree method, the satellite remote sensing image data of the pine wood nematode disease epidemic situation area is classified and extracted by utilizing the classification rule set obtained in the step S410, and pine wood nematode disease dying tree patterns in the classification and extraction result are used as pine wood nematode disease epidemic situation monitoring results aiming at the pine wood nematode disease epidemic situation area. Wherein, the step S411 further comprises steps a 1-a 3:
step a1, classifying and extracting the verification sample group by using the decision tree classification rule set to obtain a sample quality verification data result of the pine wood nematode disease dead tree;
step a2, judging according to the obtained sample quality verification data result:
when the obtained sample verification data result meets the preset precision requirement (for example, the preset precision requirement of the embodiment is 85%), the step a3 is executed; otherwise, correcting the training sample set, and executing the step S45-step S410 until the re-extracted sample verification data result meets the preset precision requirement, and executing the step a 3;
in step a3, the current decision tree classification rule set is used as the classification rule set in step S411.
Although preferred embodiments of the present invention have been described in detail hereinabove, it should be clearly understood that modifications and variations of the present invention are possible to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A remote sensing monitoring method for the planting situation of the pine wood nematode disease integrated in the sky is characterized by comprising the following steps of S1-S4:
step S1, respectively acquiring satellite remote sensing image original data of a target area to be monitored, unmanned aerial vehicle hyperspectral image original data of a pine wood nematode disease epidemic situation area and ground investigation original data of the pine wood nematode disease epidemic situation area, and forming a pine wood nematode disease epidemic situation area original data set by the acquired satellite remote sensing image original data, unmanned aerial vehicle hyperspectral image original data and the ground investigation original data; the target area to be monitored comprises a pine wood nematode disease epidemic situation area, an acquisition area corresponding to ground inspection original data is consistent with an acquisition area corresponding to unmanned aerial vehicle hyperspectral image original data, and the ground inspection original data at least comprises a position of a tree which is killed by the pine wood nematode disease in the pine wood nematode disease epidemic situation area and a picture of the tree which is killed by the pine wood nematode disease;
step S2, respectively preprocessing the original data of the satellite remote sensing image, the original data of the hyperspectral image of the unmanned aerial vehicle and the original data of the ground investigation in the original data set of the pine wood nematode disease area, correspondingly obtaining the original data of the satellite remote sensing image, the hyperspectral image of the unmanned aerial vehicle and the ground investigation data, and forming a pine wood nematode disease area data set by using the obtained data;
step S3, establishing a remote sensing interpretation sample aiming at the pine wood nematode disease epidemic situation area; the remote sensing interpretation sample comprises an image sample, a spectrum sample and sample point attribute information aiming at each type of ground object in the pine wood nematode disease situation area to be monitored;
and S4, processing the pine wood nematode disease situation monitoring result aiming at the pine wood nematode disease situation area according to the obtained pine wood nematode disease situation area data set and the remote sensing interpretation sample.
2. The remote sensing monitoring method for the skis-ground integrated pine wood nematode disease epidemic situation of claim 1, wherein in step S2, the preprocessing of the raw data of the hyperspectral image of the unmanned aerial vehicle comprises the following steps:
carrying out format conversion processing on the unmanned aerial vehicle hyperspectral image original data in the pine wood nematode disease situation area original data set; the format conversion processing comprises the steps of converting cube data of the hyperspectral image original data of the unmanned aerial vehicle into ENVI format data and converting the ENVI format data into JPG format data;
carrying out mosaic splicing treatment on the image of the original data of the hyperspectral image of the unmanned aerial vehicle after format conversion; the mosaic splicing processing comprises image alignment processing, feature point generation processing, point cloud generation processing, texture calculation processing, grid construction processing and mosaic model generation processing of the converted JPG format data.
3. The remote sensing monitoring method for the skis-ground integrated pine wood nematode disease situation of claim 2, wherein in step S2, the preprocessing of the raw data of the satellite remote sensing image comprises a geometric correction process, a radiation correction process and an atmospheric correction process aiming at the raw data of the satellite remote sensing image.
4. The remote sensing monitoring method for the skimming of pine wood nematode disease of claim 3, wherein in step S2, the preprocessing process of the ground-based raw data comprises:
inputting ground investigation original data aiming at the tree withered and dead due to the pine wilt disease in the pine wilt disease epidemic situation area;
and, carry on the standardized processing to the primitive data of this ground investigation entered; the ground-stepping raw data comprises a ground-stepping photo file path, a sample category, equipment parameters, weather parameters and remark information.
5. The remote sensing monitoring method for the skimming bulk nematode disease according to any one of claims 1-4, wherein the process of establishing the remote sensing interpretation sample comprises the following steps S31-S35:
step S31, selecting a surface feature sample set aiming at the pine wood nematode disease situation area to be monitored, and acquiring the position information and surface feature sample photos of each surface feature sample in the surface feature sample set in a surface feature examination mode; the surface feature samples in the surface feature sample set comprise healthy pine samples, dead tree samples caused by the pine wilt disease, broad-leaved tree samples, bare surface samples, building roof samples and road samples in a pine wilt disease epidemic situation area to be monitored;
step S32, on the basis of the preprocessed unmanned aerial vehicle hyperspectral image data, taking a pixel of the spatial position information of the ground stepping position point on the unmanned aerial vehicle hyperspectral image data as a center, and intercepting a rectangular range of 3 x 3 pixels as an image sample for each type of ground object in the pine wood nematode disease epidemic situation area to be monitored; wherein, the rectangular range of the 3 x 3 image elements comprises 9 image elements;
step S33, taking the intercepted average spectral characteristic curve of 9 pixels in the rectangular range of the 3 x 3 pixels as a spectral sample for each category of ground objects in the pine wood nematode disease situation area to be monitored;
and step S34, establishing a space data set of the remote sensing interpretation sample by using the space position information of the ground treading point position, endowing each ground feature sample with a unique sample code, and forming the remote sensing interpretation sample by the space position information of each ground treading point position, the image sample of each type of ground feature and the spectrum sample of each type of ground feature.
6. The remote sensing monitoring method for the pine wood nematode disease in the skied area according to claim 5, wherein in step S4, the monitoring result of the pine wood nematode disease in the pine wood nematode disease area is obtained by processing the following steps S41-S411:
step S41, superposing the pine forest class vector data in the forest resource second-class survey database with the multi-source remote sensing image by adopting a GIS superposition analysis method, and intercepting the image of the pine forest class in the pine wood nematode disease epidemic situation area to be monitored; the multisource remote sensing image is satellite remote sensing image data and unmanned aerial vehicle hyperspectral image data in the pine wood nematode disease situation area data set;
step S42, quantitative inversion extraction is carried out on hyperspectral image data of the unmanned aerial vehicle by utilizing the spectrum characteristic index model of the pine wilt disease dead tree, and a hyperspectral-based pine wilt disease dead tree extraction result is obtained;
step S43, taking the obtained pine wood nematode disease dead tree extraction result based on the hyperspectrum and other surface feature samples of the non-pine wood nematode disease dead tree sample in the remote sensing interpretation sample as area classification samples of the pine wood nematode disease epidemic situation area satellite remote sensing image data;
step S44, randomly classifying the region classification sample into two sub-region samples, taking one of the sub-region samples as a training sample group, and taking the other sub-region sample as a verification sample group;
step S45, respectively extracting spectral characteristic index sets and texture characteristic index sets of various ground features in the training sample group aiming at the satellite remote sensing image data preprocessed in the step S2, and forming training sample files by all the extracted characteristic index sets; the training sample file is marked as D, the total number of the ground feature type types in the training sample file D is v, and v is more than 1;
step S46, using the feature index corresponding to any feature index set in the training sample file as the classification attribute, and sequencing all the feature index values in the training sample file to obtain a sequenced feature index value sequence; n characteristic index sets are set in a training sample file D, and the characteristic index corresponding to any one characteristic index set is marked as An，1≤n≤N；
Step S47, respectively calculating the information gain of each characteristic index in the training sample file; wherein:
wherein p isiRepresenting the probability of the ith type ground object sample in the training sample file D; infr (D) represents the information entropy of the training sample file D;
representing characteristic indexes in training sample file DAnEntropy of the information contained, DjRepresenting characteristic index A in training sample file DnA subsample set formed when the value is j; InforGain (A)n) InforGain (A) represents characteristic index A in training sample file DnThe corresponding information gain;
step S48, calculating the splitting information degree of each characteristic index in the training sample file, and calculating the information gain rate of the characteristic index by the information gain of each characteristic index and the corresponding splitting information degree; wherein:
wherein,
indicates a characteristic index AnIs split information degree, gain ratio (A)n) GainRatio (A) represents a characteristic index AnThe information gain rate of (d);
step S49, selecting the characteristic index with the maximum information gain rate as the split node attribute from all the characteristic indexes in the training sample file, and dividing the training sample file into two sample file subsets by using the split node attribute;
s410, repeatedly processing each sample file subset according to the mode of S46-S49 until the ground feature types of the samples in each sample file subset are all single types, and taking the obtained training result as a decision tree classification rule set;
and S411, according to a decision tree method, classifying and extracting the satellite remote sensing image data of the pine wood nematode disease epidemic situation area by using the obtained classification rule set, and taking the pine wood nematode disease dying tree pattern spots in the classification extraction result as the pine wood nematode disease epidemic situation monitoring result aiming at the pine wood nematode disease epidemic situation area.
7. The remote sensing monitoring method for the skies and lands integrated pine wood nematode disease epidemic situation of claim 6, wherein in step S42, the construction process of the spectrum characteristic index model of the dead tree of pine wood nematode disease comprises the following steps b 1-b 4:
respectively removing the envelope curve of each surface feature in the surface feature sample set by adopting an envelope removal method to obtain an enhanced spectral characteristic curve of each surface feature;
extracting a difference waveband of the enhanced spectral characteristic curve of the tree dead due to the pine wilt disease in the surface feature sample set and the enhanced spectral characteristic curves of other surface features in a manual interpretation mode; wherein, the other ground features are the ground features except for the dead tree caused by the pine wilt disease in the ground feature sample set;
taking the extracted difference wave band as a characteristic wave band, and taking the characteristic wave band as spectral characteristic analysis information of the tree withered due to the pine wood nematode disease;
establishing a linear model by using the characteristic wave band, and taking the linear model as a spectral characteristic index model of the tree withered and dead due to the pine wilt disease; wherein, the spectral characteristic index of the dead tree caused by the pine wilt disease is marked as IBx：
Wherein R is526、R534And R654The spectral reflectivities at 526nm, 534nm and 654nm are hyperspectral data for the unmanned aerial vehicle.
8. The remote sensing monitoring method for the skis-ground integrated pine wood nematode disease situation of claim 7, wherein the step S411 further comprises the steps of a 1-a 3:
step a1, classifying and extracting the verification sample group by using the decision tree classification rule set to obtain a sample quality verification data result of the pine wood nematode disease dead tree;
step a2, judging according to the obtained sample quality verification data result:
when the obtained sample verification data result meets the preset precision requirement, turning to the step a 3; otherwise, correcting the training sample set, and executing the step S45-step S410 until the re-extracted sample verification data result meets the preset precision requirement, and executing the step a 3;
in step a3, the current decision tree classification rule set is used as the classification rule set in step S411.
9. The remote sensing monitoring method for the skis-ground integrated pine wood nematode disease situation according to claim 8, wherein the unmanned aerial vehicle hyperspectral image raw data is acquired by a Roc unmanned aerial vehicle with model number CW-10, and the satellite remote sensing image raw data is satellite remote sensing image data with 1m spatial resolution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110197983.2A CN113011266B (en) | 2021-02-22 | 2021-02-22 | Sky-ground integrated pine wood nematode disease epidemic situation remote sensing monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110197983.2A CN113011266B (en) | 2021-02-22 | 2021-02-22 | Sky-ground integrated pine wood nematode disease epidemic situation remote sensing monitoring method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113011266A true CN113011266A (en) | 2021-06-22 |
| CN113011266B CN113011266B (en) | 2022-09-13 |
Family
ID=76406143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110197983.2A Active CN113011266B (en) | 2021-02-22 | 2021-02-22 | Sky-ground integrated pine wood nematode disease epidemic situation remote sensing monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113011266B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114492210A (en) * | 2022-04-13 | 2022-05-13 | 潍坊绘圆地理信息有限公司 | Hyperspectral satellite borne data intelligent interpretation system and implementation method thereof |
| CN115115955A (en) * | 2022-07-08 | 2022-09-27 | 宁波大学 | Pine wood nematode disease tree monitoring method and device based on unmanned aerial vehicle hyperspectral remote sensing |
| CN116630204A (en) * | 2023-07-19 | 2023-08-22 | 南京佳格耕耘科技有限公司 | Remote sensing image online analysis processing system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360376A (en) * | 2018-09-13 | 2019-02-19 | 航天信德智图（北京）科技有限公司 | Infect the withered masson pine satellite monitoring spatial database of pine nematode |
| CN110514597A (en) * | 2019-09-04 | 2019-11-29 | 北京麦飞科技有限公司 | The diseases and pests of agronomic crop monitoring method of based on star remotely-sensed data collaboration |
| CN110533595A (en) * | 2019-09-03 | 2019-12-03 | 北京麦飞科技有限公司 | A kind of method and system of crop disease and insect Spatial distributions distribution monitoring |
| CN110865040A (en) * | 2019-11-29 | 2020-03-06 | 深圳航天智慧城市系统技术研究院有限公司 | Sky-ground integrated hyperspectral water quality monitoring and analyzing method |
-
2021
- 2021-02-22 CN CN202110197983.2A patent/CN113011266B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360376A (en) * | 2018-09-13 | 2019-02-19 | 航天信德智图（北京）科技有限公司 | Infect the withered masson pine satellite monitoring spatial database of pine nematode |
| CN110533595A (en) * | 2019-09-03 | 2019-12-03 | 北京麦飞科技有限公司 | A kind of method and system of crop disease and insect Spatial distributions distribution monitoring |
| CN110514597A (en) * | 2019-09-04 | 2019-11-29 | 北京麦飞科技有限公司 | The diseases and pests of agronomic crop monitoring method of based on star remotely-sensed data collaboration |
| CN110865040A (en) * | 2019-11-29 | 2020-03-06 | 深圳航天智慧城市系统技术研究院有限公司 | Sky-ground integrated hyperspectral water quality monitoring and analyzing method |
Non-Patent Citations (2)
| Title |
|---|
| 徐培林 等: "松材线虫病天空地一体化立体监测技术", 《测绘》 * |
| 邓世晴: "星机地协同的松材线虫病疫区枯死松树监测方法研究", 《中国优秀博硕士学位论文全文数据库（硕士） 工程科技II辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114492210A (en) * | 2022-04-13 | 2022-05-13 | 潍坊绘圆地理信息有限公司 | Hyperspectral satellite borne data intelligent interpretation system and implementation method thereof |
| CN114492210B (en) * | 2022-04-13 | 2022-07-19 | 潍坊绘圆地理信息有限公司 | Hyperspectral satellite borne data intelligent interpretation system and implementation method thereof |
| CN115115955A (en) * | 2022-07-08 | 2022-09-27 | 宁波大学 | Pine wood nematode disease tree monitoring method and device based on unmanned aerial vehicle hyperspectral remote sensing |
| CN116630204A (en) * | 2023-07-19 | 2023-08-22 | 南京佳格耕耘科技有限公司 | Remote sensing image online analysis processing system |
| CN116630204B (en) * | 2023-07-19 | 2023-09-26 | 南京佳格耕耘科技有限公司 | Remote sensing image online analysis processing system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113011266B (en) | 2022-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113011266B (en) | Sky-ground integrated pine wood nematode disease epidemic situation remote sensing monitoring method | |
| Cihlar et al. | Classification by progressive generalization: A new automated methodology for remote sensing multichannel data | |
| Eng et al. | The use of VARI, GLI, and VIgreen formulas in detecting vegetation in aerial images | |
| CN110991335A (en) | Visible light unmanned aerial vehicle remote sensing image forest tree species classification method based on multi-feature optimization | |
| CN115713694B (en) | Land mapping information management method | |
| CN106875407B (en) | Unmanned aerial vehicle image canopy segmentation method combining morphology and mark control | |
| CN113869370A (en) | Tree species refined classification method based on deep learning algorithm and time sequence sentinel image | |
| JP2011024471A (en) | Classification of kind of tree in forest by utilizing intermediate infrared area band | |
| CN112509026A (en) | Insulator crack length identification method | |
| Sathyanarayanan et al. | A multiclass deep learning approach for LULC classification of multispectral satellite images | |
| CN113033386B (en) | High-resolution remote sensing image-based transmission line channel hidden danger identification method and system | |
| CN117079059B (en) | Tree species automatic classification method based on multi-source satellite image | |
| Khalid et al. | Evaluation the accuracy of oil palm tree detection using deep learning and support vector machine classifiers | |
| CN109472244B (en) | Soil, rock and vegetation identification method based on separation index | |
| CN115810151A (en) | Deep learning woodland extraction tree species recognition algorithm model | |
| Li et al. | Tree species classification based on WorldView-2 imagery in complex urban environment | |
| CN112580504B (en) | Tree species classification counting method and device based on high-resolution satellite remote sensing image | |
| CN113705454A (en) | Method for extracting forest land containing infrared spectrum remote sensing image | |
| JP2011113233A (en) | Taxonomic image creation device | |
| CN114821370A (en) | Single-tree crown detection and segmentation method based on unmanned aerial vehicle image and U-Net | |
| Phyo et al. | Classification of Cluster Area Forsatellite Image | |
| Xiao et al. | Robust Land Cover Classification with Local-Global Information Decoupling to Address Remote Sensing Anomalous Data | |
| Idris et al. | Application of artificial neural network for building feature extraction in Abuja | |
| Deng et al. | Tree crown recognition algorithm on high spatial resolution remote sensing imagery | |
| CN113516067B (en) | Method and device for identifying plant growth time |
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 |