CN107330898A - Altitudinal vegetation zone quantitatively delineates computational methods and system - Google Patents
Altitudinal vegetation zone quantitatively delineates computational methods and system Download PDFInfo
- Publication number
- CN107330898A CN107330898A CN201710467468.5A CN201710467468A CN107330898A CN 107330898 A CN107330898 A CN 107330898A CN 201710467468 A CN201710467468 A CN 201710467468A CN 107330898 A CN107330898 A CN 107330898A
- Authority
- CN
- China
- Prior art keywords
- ndvi
- altitudinal
- vegetation
- vegetation zone
- demarcation
- 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
- 238000000205 computational method Methods 0.000 title claims abstract description 28
- 238000010586 diagram Methods 0.000 claims abstract description 56
- 238000012216 screening Methods 0.000 claims abstract description 53
- 238000000034 method Methods 0.000 claims abstract description 37
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 239000000284 extract Substances 0.000 claims abstract description 19
- 238000000605 extraction Methods 0.000 claims description 37
- 238000000926 separation method Methods 0.000 claims description 32
- 238000011160 research Methods 0.000 claims description 30
- 230000000007 visual effect Effects 0.000 claims description 24
- 238000005457 optimization Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000012937 correction Methods 0.000 claims description 15
- 230000011218 segmentation Effects 0.000 claims description 15
- 230000004927 fusion Effects 0.000 claims description 13
- 238000002474 experimental method Methods 0.000 claims description 11
- 230000005855 radiation Effects 0.000 claims description 8
- 230000007704 transition Effects 0.000 claims description 7
- 230000001373 regressive effect Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 4
- 238000003709 image segmentation Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 241001269238 Data Species 0.000 claims description 2
- 125000000058 cyclopentadienyl group Chemical group C1(=CC=CC1)* 0.000 claims 1
- ZSWFCLXCOIISFI-UHFFFAOYSA-N endo-cyclopentadiene Natural products C1C=CC=C1 ZSWFCLXCOIISFI-UHFFFAOYSA-N 0.000 claims 1
- 238000012360 testing method Methods 0.000 claims 1
- 241000220317 Rosa Species 0.000 abstract 1
- 230000008859 change Effects 0.000 description 11
- 238000011835 investigation Methods 0.000 description 6
- 241000196324 Embryophyta Species 0.000 description 5
- 230000009467 reduction Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000001556 precipitation Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 239000002689 soil Substances 0.000 description 3
- 244000025254 Cannabis sativa Species 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000001704 evaporation Methods 0.000 description 2
- 230000008020 evaporation Effects 0.000 description 2
- 238000004611 spectroscopical analysis Methods 0.000 description 2
- 238000013075 data extraction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000155 melt Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
- G06T7/41—Analysis of texture based on statistical description of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Image Processing (AREA)
Abstract
A kind of altitudinal vegetation zone quantitatively delineates computational methods and system.It the described method comprises the following steps：(1) remote sensing image data in altitudinal vegetation zone area to be extracted is obtained by satellite, the corresponding gradient and slope aspect data are extracted by this area's digital complex demodulation；(2) image procossing is carried out to the image data, and carries out sample area screening；(3) sample area scatter diagram is built, initial altitudinal vegetation zone cut off value is extracted；(4) analyzed based on Neighborhood Statistics, extract altitudinal vegetation zone line of demarcation；(5) output result image and data.The system includes data read module, constituency module, scatterplot module, neighbor analysis module and output module.The quantitative rose method of altitudinal vegetation zone described herein and system results are objective and accurate, with very high application value.
Description
Technical field
The application is related to technical field of data processing, and more particularly to altitudinal vegetation zone quantitatively delineates computational methods and is
System.
Background technology
Influenceed, vegetation pattern changes with the rising of height above sea level, presented simultaneously by factors such as temperature, precipitation, illumination, soil
Go out the feature of gradual change.At the same time, vegetation pattern is again different because of the difference of region with the change of height above sea level.These all give
Amount delineation altitudinal vegetation zone distribution characteristics has brought great challenge.
The method extracted both at home and abroad on alpine vegetation belt at present is broadly divided into three classes：Traditional ground investigation side
Method, Remote Sensing Interpretation and ecological model method.
Ground investigation method characterizes local area ecological feature using local ecological characteristic, with certain error, and there is sample
The shortcomings of point collection is difficult, time-consuming, cost is high.For alpine vegetation, these deficiencies, which then seem, to be become apparent.
Extraction is carried out to altitudinal vegetation zone using the method for remote Sensing Interpretation and is broadly divided into following six major classes method：
1) visual interpretation method
Visual interpretation method is mainly based upon high-definition remote sensing image data, altitudinal vegetation zone distributing position is carried out visual
Interpretation.This method is simple and easy to apply, but artificial subjective impact is larger.
2) land use classes method
Land use classes method is mainly based upon carries out land use classes to remote sensing image, in conjunction with visual interpretation, shape
The methods such as state, soil statistics, landscape esthetics principle carry out altitudinal vegetation zone extraction.This method depends on the essence of image classification
Degree, and it is larger to the quantitative scoring errors of altitudinal vegetation zone.
3) quantity variation
Quantity variation is to regard the pixel of remote sensing image data as sample prescription, and it is sample prescription that the spectral properties of pixel, which are regarded,
Quantative attribute, is classified using vegetation quantity classification to the pixel on line-transect.This method calculates complicated, and is only applicable to carry
Take the position in altitudinal vegetation zone boundary line on specific line-transect, it is impossible to extract a wide range of continuously distributed altitudinal vegetation zone distribution boundary.
4) Neighborhood Statistics method
Two points mainly by remote sensing image of Neighborhood Statistics method is two classes；The sizeable neighborhood window of reselection, asks for two
Divide the average value of all grids in rear remote sensing image data neighborhood；Finally definition distribution threshold value, you can quantitatively depict vegetation and hang down
Straight band line of demarcation.This method can extract various types of altitudinal vegetation zone lines of demarcation, but the definition for being distributed threshold value have compared with
Strong subjectivity.
5) edge detection method
Edge detection method is mainly classified remote sensing image data according to vegetation pattern, then classification results are divided
The steps such as class post processing, Morphological scale-space, are finally automatically identified between different vegetation types using the method for rim detection
Line of demarcation.This method is applied to extract the sudden turn of events type altitudinal vegetation zone line of demarcation of vegetation line of demarcation simple shape, and to large scale,
The relatively low image data extraction effect of spatial resolution preferably, but is not suitable for the complex-shaped gradation type vegetation in vegetation line of demarcation
Belt line of demarcation.
6) correlation analysis method correlation analysis method is mainly by analyzing normalized differential vegetation index (NDVI) and height above sea level etc.
The correlation of envirment factor carries out the position analysis of altitudinal vegetation zone line of demarcation.This method obtains geographical category first with ground surface sample
Property, the NDVI values in sampling point remote sensing image are obtained further according to geographical position, finally features of Ecological Environment, Gui Gen are analyzed using sampling point
Bottom is tied, is investigated without departing from ground sampling point.
Ecological model method is ripe not enough at present, and the factor of the modeling most critical to be considered is how to improve fitting for model
With scope and precision.When predicting altitudinal vegetation zone line of demarcation dynamic change, there is presently no preferable forecast model.
The content of the invention
The application's aims to overcome that above deficiency quantitatively delineates computational methods there is provided a kind of fine altitudinal vegetation zone
With a kind of quantitative scoring system of altitudinal vegetation zone, variation characteristics of the NDVI with height above sea level can be embodied more fully hereinafter.
The application uses following technical scheme：
Wherein, the technical scheme that a kind of fine altitudinal vegetation zone disclosed in the present application quantitatively delineates computational methods is as follows：
A kind of altitudinal vegetation zone quantitatively delineates computational methods, for quantitatively delineating altitudinal vegetation zone distribution characteristics, its feature
It is, the computational methods comprise the following steps：
(1) remote sensing image data in altitudinal vegetation zone area to be extracted is obtained by satellite, passes through this area's digital elevation
Model DEM extracts the corresponding gradient and slope aspect data；
(2) image procossing is carried out to the remote sensing image data, and carries out sample area screening；
(3) sample area scatter diagram is built, initial altitudinal vegetation zone cut off value is extracted；
(4) analyzed based on Neighborhood Statistics, extract altitudinal vegetation zone line of demarcation；
(5) output result image and data.
Altitudinal vegetation zone of the present invention quantitatively delineates computational methods and further comprises following preferred scheme：
In the step (1), remote sensing image, digital complex demodulation, the gradient and slope aspect need to be same vegetation to be extracted
Belt area.
In the step (2), carrying out image procossing to the remote sensing image data includes following content：
2.1.1 radiation calibration, atmospheric correction and topographical correction pretreatment operation are carried out to remote sensing image；
2.1.2 vegetation-cover index NDVI is extracted based on pretreated remote sensing image；
2.1.3 visual fusion is carried out to NDVI, DEM, the gradient and slope aspect data.
Wherein, vegetation-cover index NDVI calculation formula is in the step 2.1.2：
In formula, NIR and R are respectively reflectance value of the vegetation near infrared band and infrared band.
In the step (2), sample area screening includes following content：
2.2.1 the fused data to NDVI, DEM, the gradient and slope aspect carries out gradient screening, and the screening gradient is not less than 5 degree
Region is used as altitudinal vegetation zone to be extracted area；
2.2.2 slope aspect screening is carried out to altitudinal vegetation zone mountain area to be extracted, it is ensured that altitudinal vegetation zone area to be extracted is same
One slope aspect.
The step (3) includes following content：
3.1 fused datas based on the altitudinal vegetation zone to be extracted area, using digital complex demodulation with from remote sensing
The vegetation-cover index NDVI extracted in image builds DEM-NDVI scatter diagrams；
3.2 pairs of DEM-NDVI scatter diagrams carry out density slice；
3.3, based on the DEM-NDVI scatter diagrams after the density slice, specific height above sea level are obtained using moving average method
The NDVI average values and sliding average curve of height；
3.4 determine suitable matched curve function；
The variation tendency of 3.5 pairs of matched curves is analyzed, and the separation for finding out the different sections of curve extracts initial plant
By belt cut off value.
Wherein, in the step 3.2, following content is included to DEM-NDVI scatter diagrams density slice：
3.2.1 it is in the range of interval, statistical interval with 0.01 for certain specific height above sea level in DEM-NDVI scatter diagrams
Scatterplot number；
3.2.2 the NDVI density values in every 0.01 interval are calculated, calculation formula is
In formula, Mi is the NDVI density values in i-th of NDVI interval, and Ni is dissipating in i-th of NDVI interval
Points, Nmax is the maximum scatterplot number in all NDVI intervals；
3.2.3 aforesaid operations are carried out to all height above sea level in DEM-NDVI scatter diagrams, obtains DEM-NDVI density maps；
3.2.4 choose region of the NDVI density values more than 0.95 and be used as DEM-NDVI scatter diagram core spaces.
In the step 3.3, moving average method includes following content：
3.3.1 size is defined to be 100m window to ask for NDVI average values in the window ranges；
3.3.2 the window is entered into line slip by 5m sliding distance along elevation direction, so as to obtain equidistant height above sea level
NDVI average values；
3.3.3 all NDVI average points of above-mentioned acquisition are connected, you can obtain sliding average curve.
In the step 3.4, sliding average curve is fitted with n times Polynomical regressive equation；Matched curve with
The degree of fitting correlation coefficient r of DEM-NDVI sliding average curves2To weigh, N can be determined by multiple contrast experiment, choosing
Taking makes r2Reach the minimum N values of maximum as suitable matched curve N values.
In the step 3.5, the separation of curve difference section is by asking flex point and second order to lead pole matched curve function
Value point is determined.
In the step (4), based on Neighborhood Statistics analysis, extracting altitudinal vegetation zone line of demarcation includes following content：
4.1 obtain altitudinal vegetation zone line of demarcation NDVI initial thresholds and corresponding experimental threshold values scope；
4.2 are iterated optimization respectively in the range of each experimental threshold values, obtain the optimal thresholds of altitudinal vegetation zone line of demarcation NDVI
Value；
4.3 pairs of sample area NDVI images (the NDVI data i.e. in the range of sample area) carry out Optimal-threshold segmentation and (are based on
NDVI optimal thresholds carry out image segmentation to NDVI images), and is carried out reclassification and (represent that NDVI images divide with single value
Cut the different type of result) and assignment；
4.4 choose the neighborhood window of particular size, and vegetation pattern probability graph is obtained by neighbor analysis method；
4.5 define the probability distribution threshold value of different vegetation types, extract final altitudinal vegetation zone line of demarcation.
Wherein, in the step 4.1, altitudinal vegetation zone line of demarcation NDVI initial thresholds are fitted for initial altitudinal vegetation zone
NDVI values corresponding to curve difference section separation, experimental threshold values scope is to be fluctuated above and below corresponding NDVI initial thresholds
0.1 interval range；
In the step 4.2, corresponding plant is asked for respectively with maximum variance between clusters in the range of each experimental threshold values
By belt line of demarcation NDVI optimal thresholds.
Carrying out assignment to the NDVI images after Optimal-threshold segmentation in the step 4.3 includes following content：
4.3.1 the assignment scope to NDVI images after Optimal-threshold segmentation is [0,1]；
4.3.2 according to height above sea level 1 is equidistantly entered as respectively from low to high to the different classes of of NDVI images ..., 0.(such as at certain
Particular studies area, can be by the NDVI shadows after segmentation with reference to the NDVI images after original remote sensing image, height above sea level distribution characteristics and segmentation
Picture reclassification is theropencedrymion and coniferous forest, alpine scrub grassy marshland, the naked class of rock exposed soil three of high mountain, and distinguishes assignment 1,0.5 and
0.) in general, dense vegetation area can be entered as to 1, no vegetation region is entered as 0, and middle transition class vegetation region can be equidistant
It is entered as certain value in the range of [0,1].
The neighborhood window of particular size is the circular window that radius is 200m in the step 4.4.
Altitudinal vegetation zone line of demarcation is extracted in the step 4.5 includes following content：
4.5.1 vegetation pattern probability graph is based on, the isopleth at intervals of 0.1 is generated；
4.5.2 altitudinal vegetation zone line of demarcation NDVI initial thresholds are based on, the initial line of demarcation of research area's altitudinal vegetation zone is obtained；
4.5.3 by isopleth, the initial line of demarcation of altitudinal vegetation zone and the regional original remote sensing image of altitudinal vegetation zone to be extracted
Figure is overlapped, and selection matches optimal equivalence with the initial line of demarcation of altitudinal vegetation zone and remote sensing image visual interpretation vegetation pattern
Line is used as the altitudinal vegetation zone line of demarcation after optimization；
4.5.4 with reference to factors such as theoretical distribution height above sea level, the slope aspects of vegetation, the altitudinal vegetation zone line of demarcation after optimization is carried out
Delete, obtain final altitudinal vegetation zone line of demarcation.
The application further simultaneously discloses a kind of quantitative scoring system of altitudinal vegetation zone, and the technical scheme of the invention is as follows：
A kind of quantitative scoring system of altitudinal vegetation zone, for quantitatively delineating altitudinal vegetation zone distribution characteristics, the vegetation is hung down
The straight quantitative scoring system of band includes data read module, constituency module, scatterplot module, neighbor analysis module and output module；
It is characterized in that：
The data read module is used for remote sensing image, the digital elevation model for reading altitudinal vegetation zone area to be extracted
DEM, image procossing is carried out to the remote sensing image that is read, extract after pretreatment the vegetation-cover index of remote sensing image, the gradient with
Slope aspect data, and the fused data of NDVI, DEM, the gradient and slope aspect is provided to constituency module；
The constituency module to being handled through data read module after remote sensing image carry out the gradient screening with slope aspect subregion, from
And altitudinal vegetation zone area to be extracted is obtained as research area, and it is vertical to the vegetation to be extracted after the offer screening of scatterplot module
Band mountain area fused data；
The scatterplot module is built according to the altitudinal vegetation zone mountain area to be extracted fused data after the screening of constituency module
DEM-NDVI scatter diagrams, obtain altitudinal vegetation zone NDVI and DEM critical value, and provide altitudinal vegetation zone to neighbor analysis module
Line of demarcation NDVI initial thresholds；
The neighbor analysis module, for obtaining altitudinal vegetation zone line of demarcation, and provides to output module final vegetation
Belt line of demarcation vector data and NDVI and the critical average values of DEM；
The output module, for exporting altitudinal vegetation zone line of demarcation vector data and NDVI and the critical average values of DEM.
The quantitative scoring system of altitudinal vegetation zone of the present invention further comprises following preferred scheme：
In the data read module, remote sensing image, DEM, the gradient and slope aspect need to be same research area scope.
The data read module includes image pre-processing unit, NDVI extraction units and visual fusion unit：
Wherein, described image pretreatment unit, for carrying out radiation calibration, atmospheric correction and landform school to remote sensing image
Positive pretreatment operation；
The NDVI extraction units are used for the vegetation-cover index NDVI for extracting remote sensing image after pretreatment；
The visual fusion unit is used to carry out visual fusion to NDVI, DEM, the gradient and slope aspect data.
The NDVI is vegetation index, and calculation formula is：
In formula, NIR and R are respectively reflectance value of the vegetation near infrared band and infrared band.
The constituency module includes gradient screening unit and slope aspect screening unit：
The gradient screening unit is used to carry out the fused data of NDVI, DEM, the gradient and slope aspect gradient screening, screening
Region of the gradient not less than 5 degree is used as altitudinal vegetation zone mountain area to be extracted；
The slope aspect screening unit, for carrying out slope aspect screening to altitudinal vegetation zone mountain area to be extracted, it is ensured that plant to be extracted
It is same slope aspect by belt mountain area.
The scatterplot module includes scatter diagram generation unit, density slice unit, moving average unit, fitting song
Line unit and separation extraction unit：
Wherein, the scatter diagram generation unit is used to read the altitudinal vegetation zone mountain to be extracted after the module screening of constituency
Area's fused data, and build DEM-NDVI scatter diagrams using DEM and NDVI；
The density slice unit is used to carry out density slice to the DEM-NDVI scatter diagrams；
The moving average unit is used to carry out moving average to the DEM-NDVI scatter diagrams after the density slice,
And obtain the NDVI average values and sliding average curve of specific height above sea level；
The matched curve unit is used to determine suitable matched curve function；
The separation extraction unit is used to analyze the variation tendency of matched curve, finds out the different sections of curve
Separation.
The density slice unit includes scatterplot statistics subelement, density computation subunit, density map subelement and core
Area generates subelement；
Wherein, the scatterplot statistics subelement is used for the scatterplot for reading the specific height above sea level of certain in DEM-NDVI scatter diagrams, with
NDVI values 0.01 are interval, count the scatterplot number in certain specific height above sea level interval；
The density computation subunit is used to calculate the NDVI density values in 0.01 interval, and calculation formula is
In formula, Mi is the NDVI density values in i-th of NDVI interval, and Ni is dissipating in i-th of NDVI interval
Points, Nmax is the maximum scatterplot number in all NDVI intervals；
The density map subelement is used to carry out aforesaid operations to all height above sea level in DEM-NDVI scatter diagrams, obtains DEM-
NDVI density maps；
The core space generation subelement is used to choose region of the NDVI density values more than 0.95 as DEM-NDVI scatterplots
Distribution map core space.
The moving average unit includes window mean value computation subelement, equidistant mean value computation subelement and curve generation
Subelement；
Wherein, the window mean value computation subelement is used to define size for 100m window to ask in the window ranges
NDVI average values；
The equidistant mean value computation subelement is used to the window entering line slip along elevation direction by 5m sliding distance,
Obtain the NDVI average values of equidistant height above sea level；
The curve generates are used to all NDVI average points of above-mentioned acquisition being connected, and obtain sliding average bent
Line.
In matched curve unit, sliding average curve is carried out curve fitting using n times Polynomical regressive equation, and
By matched curve and the degree of fitting correlation coefficient r of DEM-NDVI sliding average curves2Weighed, wherein, N passes through multiple
Contrast experiment determines that selection makes r2Reach the minimum N values of maximum as suitable matched curve N values.
In separation extraction unit, determine that fitting is bent by asking flex point and second order to lead extreme point matched curve function
The separation of line difference section.
The neighbor analysis module includes threshold range acquiring unit, optimal threshold extraction unit, reclassification unit, probability
Figure generation unit and boundary straight line extraction unit；
Wherein, the threshold range acquiring unit is used to obtain altitudinal vegetation zone line of demarcation NDVI initial thresholds and corresponding
Experimental threshold values scope；
The optimal threshold extraction unit is used to be iterated optimization respectively in the range of each experimental threshold values, obtains vegetation and hangs down
Straight band line of demarcation NDVI optimal thresholds；
The reclassification unit is used to carry out Optimal-threshold segmentation to the NDVI data in the range of sample area, and is carried out
Reclassification and assignment；
The probability graph generation unit is used for the neighborhood window for choosing particular size, and vegetation class is obtained by neighbor analysis method
Type probability graph；
The boundary straight line extraction unit, the probability distribution threshold value for defining different vegetation types extracts final vegetation
Belt line of demarcation.
It is in threshold range acquiring unit, matched curve determined by separation extraction unit in scatterplot module is different
NDVI values corresponding to section separation are as altitudinal vegetation zone line of demarcation NDVI initial thresholds, by corresponding NDVI initial thresholds
The interval range of fluctuation 0.1 is used as experimental threshold values scope up and down.
The optimal threshold extraction unit includes inter-class variance subelement, and the inter-class variance subelement is used in sample area
In the range of ask for corresponding vegetation respectively with the thought of maximum variance between clusters in the range of each experimental threshold values of NDVI images and hang down
Straight band line of demarcation NDVI optimal thresholds.
The reclassification unit is used to determine assignment scope and the sorted image of counterweight carries out assignment.
In the probability graph generation unit, the neighborhood window of particular size is the circular window that radius is 200m.
At the beginning of the boundary straight line extraction unit includes isopleth generation subelement, initial line of demarcation acquisition subelement, line of demarcation
Optimization subelement at the beginning of optimizing subelement, line of demarcation and line of demarcation essence extract subelement；
Wherein, isopleth generation subelement is used for based on the vegetation pattern probability graph acquired in probability graph generation unit, raw
At interval be 0.1 isopleth；
The initial line of demarcation, which obtains subelement, to be used to be based on altitudinal vegetation zone line of demarcation NDVI initial thresholds, obtains research
The initial line of demarcation of area's altitudinal vegetation zone；
Optimize subelement at the beginning of the line of demarcation for isopleth, the initial line of demarcation of altitudinal vegetation zone and research area is original distant
Sense striograph is overlapped, and selection is matched most preferably with the initial line of demarcation of altitudinal vegetation zone and remote sensing image visual interpretation vegetation pattern
Isopleth be used as the altitudinal vegetation zone line of demarcation after optimization；
The line of demarcation essence, which extracts subelement, to be used to combine the factors such as theoretical distribution height above sea level, the slope aspect of vegetation, after optimization
Altitudinal vegetation zone line of demarcation deleted, obtain final altitudinal vegetation zone line of demarcation.
Altitudinal vegetation zone disclosed in the present application, which quantitatively delineates algorithm and system, has advantages below：
1) compared with traditional ground investigation method, the application is used as sample point, a side using pixels all in area are studied
Face adds the quantity of sample point, sample is preferably characterized the overall distribution situation of altitudinal vegetation zone；On the other hand, reduce
The time of ground investigation and human input；Screened in addition, the application has carried out the gradient to research area with slope aspect, it is ensured that experiment sample
Local area can more realistically characterize the distribution characteristics of research area's altitudinal vegetation zone, make quantitative delineation result more accurately and reliably.
2) compared with current existing remote Sensing Interpretation method, the altitudinal vegetation zone line of demarcation extracted be based on trend fitting with
Rule is counted so that altitudinal vegetation zone distributed architecture is more regular, reduces the influence of artificial subjectivity.
3) before carrying out curve fitting, the application has carried out density slice to scatter diagram, effectively eliminates in sample value
Random fluctuation.
4) present invention asks for corresponding plant respectively in the range of each experimental threshold values with the thought of maximum variance between clusters
By belt line of demarcation NDVI optimal thresholds, the amount of calculation of iteration optimization is on the one hand reduced；On the other hand noise is decreased
Influence, weaken the limitation of maximum variance between clusters；In addition, also making extracted altitudinal vegetation zone line of demarcation more smart
It is accurate.
Brief description of the drawings
Fig. 1 is the schematic flow sheet that altitudinal vegetation zone disclosed in the present application quantitatively delineates computational methods；
Fig. 2 is the structural representation of the quantitative scoring system of altitudinal vegetation zone disclosed in the present application.
Embodiment
As shown in Figure 1, computational methods are quantitatively delineated this application discloses a kind of altitudinal vegetation zone, for quantitatively delineating plant
By belt distribution characteristics, comprise the following steps：
Step 1：The remote sensing image data in altitudinal vegetation zone area to be extracted is obtained by satellite, it is high by this area's numeral
Journey model DEM extracts the corresponding gradient and slope aspect data.
In the present embodiment, remote sensing image, DEM, the gradient and slope aspect need to be same research area scope.To ensure that experimental data is equal
For same research area scope, original remote sensing image and DEM need to be cut with research area's vector data.The gradient and slope aspect data
It is the DEM extraction results after cutting.
Step 2：Image procossing is carried out to the image data, and carries out sample area screening.
In the present embodiment, above-mentioned steps 2 are comprised the following specific steps that：
Step 2.1：Image procossing is carried out to the remote sensing image data；
Step 2.2：Sample area screening is carried out to the remote sensing image after image procossing.
Wherein, step 2.1 specifically includes below scheme：
Step 2.1.1：Radiation calibration, atmospheric correction and topographical correction pretreatment operation are carried out to remote sensing image；
Because sensor information is uneven, sensor parameters change, hypsography, the shadow such as earth rotation and atmospheric refraction
Ring, remote sensing image can be made to deform error, while spectroscopic data also can distortion.Therefore, needed before using remote sensing image data
The pretreatment operations such as radiation calibration, atmospheric correction, topographical correction, visual fusion, image enhaucament are carried out to remote sensing image.
Step 2.1.2：Vegetation-cover index NDVI is extracted based on pretreated remote sensing image.NDVI calculation formula are：
In formula, NIR and R are respectively reflectance value of the vegetation near infrared band and infrared band.
Step 2.1.3：Visual fusion is carried out to NDVI, DEM, the gradient and slope aspect data.
Step 2.2 includes herein below：
Step 2.2.1：Gradient screening is carried out to fused data, region of the screening gradient not less than 5 degree is planted as to be extracted
By belt mountain area；
Step 2.2.2：Slope aspect screening is carried out to altitudinal vegetation zone mountain area to be extracted, it is ensured that altitudinal vegetation zone mountain area to be extracted
For same slope aspect.
The present embodiment requires Experimental Area for single hillside, and constituency natural transition from mountain bottom to mountain top, and each vegetation is vertical
Band is uniformly distributed.And the mountain region gradient can be divided into following several classes：Flat slope (less than 5 °), gentle slope (5-15 °), slope (16-25 °),
Abrupt slope (26-35 °), steep slope (36-45 °) and dangerous slope (more than 45 °) etc..Therefore, when carrying out sample area screening, on the one hand require
The gradient is not less than 5 degree, on the other hand requires to be same slope aspect.
Step 3：Sample area scatter diagram is built, initial altitudinal vegetation zone cut off value is extracted.
In the present embodiment, above-mentioned steps 3 are comprised the following specific steps that：
Step 3.1：Based on the altitudinal vegetation zone mountain area to be extracted fused data, DEM- is built using DEM and NDVI
NDVI scatter diagrams；
Step 3.2：Density slice is carried out to the DEM-NDVI scatter diagrams；
In Different Altitude, DEM has different distribution characteristics from NDVI.In height above sea level, lower ground altitudinal vegetation zone is not present in area
Line of demarcation, NDVI values are of a relatively high；In the higher mountain top part of height above sea level completely without vegetation, NDVI values are relatively low；On the hillside
Place, NDVI values are raised and reduced with height above sea level.In general, NDVI meets parabolical trend with the matched curve of altitude change,
But unavoidably there is certain random fluctuation.In the present embodiment, DEM-NDVI scatter diagrams are visually observed, finds mesh
The NDVI spans of scatter diagram core space are regarded as 0.2, therefore are scatterplot number in the range of interval, statistical interval with 0.01, scatterplot is obtained
Density map, passes through the screening to density value, you can obtain the core space of DEM-NDVI scatter diagrams.So can effectively it eliminate
Random fluctuation in sample value.
In the present embodiment, density slice includes following content：
3.2.1 it is interval, statistical interval model with NDVI values 0.01 for certain specific height above sea level in DEM-NDVI scatter diagrams
Enclose interior scatterplot number；
3.2.2 the NDVI density values in 0.01 interval are calculated, calculation formula is
In formula, Mi is the NDVI density values in i-th of NDVI interval, and Ni is dissipating in i-th of NDVI interval
Points, Nmax is the maximum scatterplot number in all NDVI intervals；
3.2.3 aforesaid operations are carried out to all height above sea level in DEM-NDVI scatter diagrams, obtains DEM-NDVI density maps；
3.2.4 choose region of the NDVI density values more than 0.95 and be used as DEM-NDVI scatter diagram core spaces.
Step 3.3：Based on the DEM-NDVI scatter diagrams after the density slice, obtain specific using moving average method
The NDVI average values and sliding average curve of height above sea level；
In the present embodiment, moving average method includes following content：
3.3.1 size is defined to be 100m window to ask for NDVI average values in the window ranges；
3.3.2 the window is entered into line slip by 5m sliding distance along elevation direction, so as to obtain equidistant height above sea level
NDVI average values；
3.3.3 all NDVI average points of above-mentioned acquisition are connected, you can obtain sliding average curve.
In the present embodiment, vegetation pattern shows vertical zonality feature, this vertical zone with the rise of height above sea level
Property has certain fluctuation, and moving average method can effectively eliminate the random fluctuation in observation.
To determine influence of the different size of sliding window to experimental precision, different window is carried out respectively to DEM-NDVI scatterplots
The experiment of mouth size, finds window size when within 100m, the sliding average plots changes base of different windows size
This is consistent, only otherwise varied in terms of local curve fluctuation, and window is smaller, and curve fluctuation is bigger.When window size is
During 100m, curve is the most smooth, and the curve of curve and window size within 100m is completely superposed, and variation tendency is completely the same.
When sliding window is more than 100m (such as 200m), curve remains unchanged smoothly, but the variation tendency of curve has deviateed window size and is
Sliding average curve within 100m.Therefore, it is that 100m window is averaged to ask for NDVI in the window ranges to define size
Value.
To determine influence of the different sliding distances to experimental precision, different sliding distances are carried out respectively to DEM-NDVI scatterplots
Experiment, find when sliding distance be more than 20m when, curve of sliding start to High aititude region translate.In theory, sliding distance is got over
Small, resulting sliding average curve is more consistent with actual conditions.To ensure precision, sliding window is set to 100m,
Sliding distance is set to DEM vertical precisions (20m) a quarter, that is, asks for the sliding window of NDVI average values high for 100m
Journey scope, each sliding distance is 5m.Such as using average value being averaged as height above sea level 3050m of NDVI in 3000~3100m of height above sea level
NDVI, slide 5m after, using the average value of NDVI in 3005~3105m of height above sea level as height above sea level 3055m average NDVI.
Step 3.4：It is determined that suitable matched curve function；
In the present embodiment, sliding average curve can be fitted with n times Polynomical regressive equation.Matched curve and DEM-
The degree of fitting of NDVI sliding average curves can use correlation coefficient r2To weigh, N can be determined by multiple contrast experiment.Choose
Make r2Reach the minimum N values of maximum as suitable matched curve N values.
Step 3.5：Variation tendency to matched curve is analyzed, and finds out the separation of the different sections of curve.
Same vegetation pattern, its NDVI has similitude (uniformity) with the variation tendency of height above sea level, in matched curve i.e.
Show as with approximate slope and slope variation speed；Transition region between different vegetation, vegetation structure is relative complex, thus
Its NDVI is relatively unstable with the variation tendency of height above sea level, is shown as with fluctuation larger slope and slope in matched curve
Rate of change.In low altitude area region, NDVI fluctuating ranges are smaller, and after certain height above sea level is reached, NDVI drastically diminishes.Cause
This, the separation of curve difference section is determined by asking flex point and second order to lead extreme point matched curve function.For
Same type of vegetation, its NDVI fall off rate is closer to.And the separation of different vegetation types should be NDVI in theory
Fall off rate most fast point.
In the present embodiment, the interval matched curve of relatively low height above sea level, approximate straight line, the single order in the region is led (i.e. tiltedly
Rate) only have a little fluctuation, therefore can be considered the approximate area of NDVI fall off rates, i.e., vegetation region of the same race.When height above sea level rises to certain specific height above sea level A
After above, NDVI fall off rates start to accelerate, i.e., internal vegetation pattern changes, therefore judges the corresponding NDVI values in A places
It can be considered a line of demarcation NDVI value of altitudinal vegetation zone.After height above sea level rises to B by A, NDVI, which remains unchanged, to diminish, but its lower reduction of speed
The variation tendency of rate starts to slow down, because the separation of different vegetation types should be that NDVI fall off rates are most fast in theory
Point, therefore judge that the corresponding NDVI values in B places can be considered another line of demarcation NDVI value of altitudinal vegetation zone.When height above sea level is risen to by B
Second dervative reaches that maximum, i.e. NDVI fall off rates reach that hereafter NDVI fall off rates start slack-off most soon at C, C.More than C
Height above sea level it is interval, NDVI fall off rates tend towards stability again, and its first derivative fluctuating range is approximate with relatively low height above sea level interval, belongs to small
Amplitude fluctuation, therefore judge that the corresponding NDVI values in C places can be considered the another bar line of demarcation NDVI values of altitudinal vegetation zone.Thus
Find out all separations of the different sections of matched curve.
Step 4：Based on Neighborhood Statistics analysis, altitudinal vegetation zone line of demarcation is extracted.
In the present embodiment, above-mentioned steps 4 are comprised the following specific steps that：
Step 4.1：Obtain altitudinal vegetation zone line of demarcation NDVI initial thresholds and corresponding experimental threshold values scope；
In the present embodiment, altitudinal vegetation zone line of demarcation NDVI initial thresholds are the different sections point of curve in the step 3.5
NDVI values corresponding to boundary's point, experimental threshold values scope is the interval range of fluctuation 0.1 above and below corresponding NDVI initial thresholds；
Step 4.2：It is iterated optimization respectively in the range of each experimental threshold values, obtains altitudinal vegetation zone line of demarcation NDVI most
Good threshold value；
In the present embodiment, based on research area's NDVI images, with maximum variance between clusters in the range of each experimental threshold values
Thought asks for corresponding altitudinal vegetation zone line of demarcation NDVI optimal thresholds respectively.
The basic thought of maximum variance between clusters is to split data into two classes using a certain threshold value, and selection makes side between two classes
Poor maximum threshold value is used as optimal threshold.When the area of two classes is more or less the same, effectively image can be split；But when two
When class area difference is larger, segmentation effect is then poor.
In the present embodiment, to avoid the disadvantages described above of maximum between-cluster variance, selection is used in the range of each experimental threshold values
The thought of maximum variance between clusters asks for corresponding optimal threshold respectively, can not only so reduce amount of calculation, also can guarantee that adjacent
The area of two classes is more or less the same, so as to more effectively split to image.In addition, initial in altitudinal vegetation zone line of demarcation NDVI
Optimal threshold extraction is carried out on the basis of threshold value, can also make final quantitative delineation result more accurate.
Step 4.3：Optimal-threshold segmentation is carried out to research area's NDVI images, and is carried out reclassification and assignment；
In the present embodiment, assignment scope is [0,1], and the different classes of of image is equidistantly assigned respectively from low to high according to height above sea level
It is worth for 1,0.6,0.3,0.
Step 4.4：The neighborhood window of particular size is chosen, vegetation pattern probability graph is obtained by neighbor analysis method；
In the present embodiment, the neighborhood window of particular size is the circular window that radius is 200m.
Neighborhood Statistics are that centered on grid to be calculated, the grid point value of certain limit around it (field) is counted
Analysis, is then output to the grid positions by end value and forms new raster map layer.The replacement of altitudinal vegetation zone mainly due to
Caused by the change of altitudinal gradient, grid is with respect to homogeneity in belt, with the rise of height above sea level, and heterogeneous grid gradually increases,
Until transition is another altitudinal vegetation zone type.Vegetation pattern probability graph progress Neighborhood Statistics are asked with the average value in neighborhood, then
Grid property value difference in same altitudinal vegetation zone is smaller, and grid property value should be close to two at different altitudinal vegetation zones
The average of altitudinal vegetation zone., then can not be by between different vegetation types when the field radius of use is too small compared with intermediate zone
Difference increase and protrude boundary line；And when the field radius of use is excessive, then this band vegetation pattern that some can be caused larger
Patch is smoothed processing, while also resulting in the reduction of Neighborhood Statistics figure resolution ratio.
In the present embodiment, by being tested to the different size of radius of neighbourhood, find when neighborhood window is that radius is
During 200m circular window, the pixel property value of same kind vegetation generally changes without too big, but some less non-
The patch of the zone of vegetation is smoothed, and the contrast between each altitudinal vegetation zone becomes apparent.
Step 4.5：The probability distribution threshold value of different vegetation types is defined, final altitudinal vegetation zone line of demarcation is extracted.
In the present embodiment, extracting altitudinal vegetation zone line of demarcation includes following content：
4.5.1 based on the vegetation pattern probability graph acquired in step 4.4, the isopleth at intervals of 0.1 is generated；
4.5.2 altitudinal vegetation zone line of demarcation NDVI initial thresholds are based on, the initial line of demarcation of research area's altitudinal vegetation zone is obtained；
4.5.3 isopleth, the initial line of demarcation of altitudinal vegetation zone and research area's original remote sensing image figure are overlapped, chosen
After optimal isopleth being matched with the initial line of demarcation of altitudinal vegetation zone and remote sensing image visual interpretation vegetation pattern as optimization
Altitudinal vegetation zone line of demarcation；
4.5.4 with reference to factors such as theoretical distribution height above sea level, the slope aspects of vegetation, the altitudinal vegetation zone line of demarcation after optimization is carried out
Delete, obtain final altitudinal vegetation zone line of demarcation.
In the present embodiment, Neighborhood Statistics are split according only to vegetation distribution density to extract altitudinal vegetation zone point with isopleth
Boundary line, and according to ground investigation on the spot, research area's north slope (Schattenseite) evaporation capacity is small, moisture condition is good, and vegetation growing way is vigorous, Nan Po
For the less leeward slope of precipitation, the landscape of " northern woods south grass " is formed on the contrary.Therefore, vegetation is extracted according only to vegetation distribution density
Belt line of demarcation can have larger erroneous judgement.When quantitatively delineating research area progress altitudinal vegetation zone, also need to combine vegetation
The factor such as theoretical distribution height above sea level, slope aspect, the isopleth acquired in step 4.5.3 is judged and deleted, obtains final
Altitudinal vegetation zone line of demarcation.
Step 5：Output result image and data.
As shown in Figure 2, disclosed herein as well is a kind of quantitative scoring system of altitudinal vegetation zone, for quantitatively delineating vegetation
Belt distribution characteristics, including data read module, constituency module, scatterplot module, neighbor analysis module and output module.
Data read module is mainly used in digital independent and image preliminary treatment, and to constituency module provide NDVI, DEM,
The fused data of the gradient and slope aspect；Constituency module mainly carries out gradient screening and slope aspect subregion to fused data, and to scatterplot
Module provides the altitudinal vegetation zone mountain area to be extracted fused data after screening；Scatterplot module is mainly used in building DEM-NDVI
Scatter diagram, and provide altitudinal vegetation zone line of demarcation NDVI initial thresholds to neighbor analysis module；Neighbor analysis module, is mainly used in
Obtain altitudinal vegetation zone line of demarcation, and to output module provide final altitudinal vegetation zone line of demarcation vector data and NDVI with
The critical average values of DEM；Output module, then be mainly used in output altitudinal vegetation zone line of demarcation vector data and NDVI and DEM faces
Boundary's average value.
Module 1：Data read module, for reading remote sensing image, digital complex demodulation, the gradient and slope aspect data, and
Carry out image procossing.
In the present embodiment, remote sensing image, DEM, the gradient and slope aspect need to be same research area scope.To ensure that experimental data is equal
For same research area scope, original remote sensing image and DEM need to be cut with research area's vector data.The gradient and slope aspect data
It is the DEM extraction results after cutting.
Above-mentioned module (1) includes：
1.1 image pre-processing units, locate in advance for carrying out radiation calibration, atmospheric correction and topographical correction to remote sensing image
Reason operation；
1.2NDVI extraction units, the vegetation-cover index NDVI for extracting remote sensing image after pretreatment,
1.3 visual fusion units, for carrying out visual fusion to NDVI, DEM, the gradient and slope aspect data.
In said units 1.2, NDVI is vegetation index, and calculation formula is：
In formula, NIR and R are respectively reflectance value of the vegetation near infrared band and infrared band.
In the present embodiment, because sensor information is uneven, sensor parameters change, hypsography, earth rotation and greatly
The influences such as gas refraction, can make remote sensing image deform error, while spectroscopic data also can distortion.Therefore, remote sensing shadow is being used
As needing that the pretreatment such as radiation calibration, atmospheric correction, topographical correction, visual fusion, image enhaucament is carried out to remote sensing image before data
Operation.
Module 2：Constituency module, for carrying out gradient screening and slope aspect subregion to research area.
In the present embodiment, above-mentioned constituency module includes gradient screening unit and slope aspect screening unit.Gradient screening unit, is used
In carrying out gradient screening to the fused data of NDVI, DEM, the gradient and slope aspect, region of the screening gradient not less than 5 degree is as waiting to carry
Take altitudinal vegetation zone mountain area；Slope aspect screening unit, for carrying out slope aspect screening to altitudinal vegetation zone mountain area to be extracted, it is ensured that wait to carry
It is same slope aspect to take altitudinal vegetation zone mountain area.
In constituency module, gradient screening is system automatic screening, and slope aspect subregion is semi-automatic subregion.By being manually entered N
Individual core slope aspect (the present embodiment is the southeast and northwest slope aspect), system then splits data into N section automatically, is subsequently grasped respectively
Make and export corresponding result.
Module 3：Scatterplot module, for building DEM-NDVI scatter diagrams, and obtains facing for altitudinal vegetation zone NDVI and DEM
Dividing value.
In the present embodiment, above-mentioned scatterplot module includes such as lower unit：
Unit 3.1：Scatter diagram generation unit, melts for reading the altitudinal vegetation zone mountain area to be extracted after the screening
Data are closed, and DEM-NDVI scatter diagrams are built using DEM and NDVI；
Unit 3.2：Density slice unit, for carrying out density slice to the DEM-NDVI scatter diagrams；
In the present embodiment, unit 3.2 includes following subelement：
Subelement 3.2.1：Scatterplot statistics subelement, the scatterplot for reading the specific height above sea level of certain in DEM-NDVI scatter diagrams,
It is interval with 0.01, counts the scatterplot number in certain specific height above sea level interval；
Subelement 3.2.2：Density computation subunit, for calculating the NDVI density values in the interval of NDVI values 0.01,
Calculation formula is
In formula, Mi is the NDVI density values in i-th of NDVI interval, and Ni is dissipating in i-th of NDVI interval
Points, Nmax is the maximum scatterplot number in all NDVI intervals；
Subelement 3.2.3：Density map subelement, for carrying out above-mentioned behaviour to all height above sea level in DEM-NDVI scatter diagrams
Make, obtain DEM-NDVI density maps；
Subelement 3.2.4：Core space generates subelement, for choosing region of the NDVI density values more than 0.95 as DEM-
NDVI scatter diagram core spaces.
In the present embodiment, in Different Altitude, DEM has different distribution characteristics from NDVI.In height above sea level, lower ground area is not
There is altitudinal vegetation zone line of demarcation, NDVI values are of a relatively high；In the higher mountain top part of height above sea level completely without vegetation, NDVI value phases
To relatively low；Locate on the hillside, NDVI values are raised and reduced with height above sea level.In general, NDVI meets with the matched curve of altitude change
Parabolical trend, but unavoidably there is certain random fluctuation.In the present embodiment, mesh is carried out to DEM-NDVI scatter diagrams
Depending on observation, the NDVI spans for finding visual scatter diagram core space are 0.2, therefore are interval, statistical interval scope with NDVI values 0.01
Interior scatterplot number, obtains scatterplot density map, passes through the screening to density value, you can obtain the core of DEM-NDVI scatter diagrams
Area.It so can effectively eliminate the random fluctuation in sample value.
Unit 3.3：Moving average unit, for entering line slip to the DEM-NDVI scatter diagrams after the density slice
It is average, and obtain the NDVI average values and sliding average curve of specific height above sea level；
In the present embodiment, unit 3.3 includes following subelement：
Subelement 3.3.1：Window mean value computation subelement, the window model is asked for for defining size for 100m window
Enclose interior NDVI average values；
Subelement 3.3.2：Equidistant mean value computation subelement, for by the window by 5m sliding distance along elevation direction
Enter line slip, obtain the NDVI average values of equidistant height above sea level；
Subelement 3.3.3：Curve generates, for all NDVI average points of above-mentioned acquisition to be connected, obtain and slide
Dynamic average value curve.
In the present embodiment, vegetation pattern shows vertical zonality feature, this vertical zone with the rise of height above sea level
Property has certain fluctuation, and moving average method can effectively eliminate the random fluctuation in observation.
To determine influence of the different size of sliding window to experimental precision, different window is carried out respectively to DEM-NDVI scatterplots
The experiment of mouth size, finds window size when within 100m, the sliding average plots changes base of different windows size
This is consistent, only otherwise varied in terms of local curve fluctuation, and window is smaller, and curve fluctuation is bigger.When window size is
During 100m, curve is the most smooth, and the curve of curve and window size within 100m is completely superposed, and variation tendency is completely the same.
When sliding window is more than 100m (such as 200m), curve remains unchanged smoothly, but the variation tendency of curve has deviateed window size and is
Sliding average curve within 100m.Therefore, it is that 100m window is averaged to ask for NDVI in the window ranges to define size
Value.
To determine influence of the different sliding distances to experimental precision, different sliding distances are carried out respectively to DEM-NDVI scatterplots
Experiment, find when sliding distance be more than 20m when, curve of sliding start to High aititude region translate.In theory, sliding distance is got over
Small, resulting sliding average curve is more consistent with actual conditions.To ensure precision, sliding window is set to 100m,
Sliding distance is set to DEM vertical precisions (20m) a quarter, that is, asks for the sliding window of NDVI average values high for 100m
Journey scope, each sliding distance is 5m.Such as using average value being averaged as height above sea level 3050m of NDVI in 3000~3100m of height above sea level
NDVI, slide 5m after, using the average value of NDVI in 3005~3105m of height above sea level as height above sea level 3055m average NDVI.
Unit 3.4：Matched curve unit, for determining suitable matched curve function；
In the present embodiment, sliding average curve can be fitted with n times Polynomical regressive equation.Matched curve and DEM-
The degree of fitting of NDVI sliding average curves can use correlation coefficient r2To weigh, N can be determined by multiple contrast experiment.Choose
Make r2Reach the minimum N values of maximum as suitable matched curve N values.
Unit 3.5：Separation extraction unit, is analyzed for the variation tendency to matched curve, finds out curve different
The separation of section.
Same vegetation pattern, its NDVI has similitude (uniformity) with the variation tendency of height above sea level, in matched curve i.e.
Show as with approximate slope and slope variation speed；Transition region between different vegetation, vegetation structure is relative complex, thus
Its NDVI is relatively unstable with the variation tendency of height above sea level, is shown as with fluctuation larger slope and slope in matched curve
Rate of change.In low altitude area region, NDVI fluctuating ranges are smaller, and after certain height above sea level is reached, NDVI drastically diminishes.Cause
This, the separation of curve difference section is determined by asking flex point and second order to lead extreme point matched curve function.For
Same type of vegetation, its NDVI fall off rate is closer to.And the separation of different vegetation types should be NDVI in theory
Fall off rate most fast point.
In the present embodiment, the interval matched curve of relatively low height above sea level, approximate straight line, the single order in the region is led (i.e. tiltedly
Rate) only have a little fluctuation, therefore can be considered the approximate area of NDVI fall off rates, i.e., vegetation region of the same race.When height above sea level rises to certain specific height above sea level A
After above, NDVI fall off rates start to accelerate, i.e., internal vegetation pattern changes, therefore judges the corresponding NDVI values in A places
It can be considered a line of demarcation NDVI value of altitudinal vegetation zone.After height above sea level rises to B by A, NDVI, which remains unchanged, to diminish, but its lower reduction of speed
The variation tendency of rate starts to slow down, because the separation of different vegetation types should be that NDVI fall off rates are most fast in theory
Point, therefore judge that the corresponding NDVI values in B places can be considered another line of demarcation NDVI value of altitudinal vegetation zone.When height above sea level is risen to by B
Second dervative reaches that maximum, i.e. NDVI fall off rates reach that hereafter NDVI fall off rates start slack-off most soon at C, C.More than C
Height above sea level it is interval, NDVI fall off rates tend towards stability again, and its first derivative fluctuating range is approximate with relatively low height above sea level interval, belongs to small
Amplitude fluctuation, therefore judge that the corresponding NDVI values in C places can be considered the another bar line of demarcation NDVI values of altitudinal vegetation zone.Thus
Find out all separations of the different sections of matched curve.
Module 4：Neighbor analysis module, for obtaining altitudinal vegetation zone line of demarcation
In the present embodiment, above-mentioned neighbor analysis module includes such as lower unit：
Unit 4.1：Threshold range acquiring unit, for obtaining altitudinal vegetation zone line of demarcation NDVI initial thresholds and corresponding
Experimental threshold values scope；
In the present embodiment, altitudinal vegetation zone line of demarcation NDVI initial thresholds are curve in the 22.5 separation extraction unit
NDVI values corresponding to different section separations, experimental threshold values scope is that 0.1 is fluctuated above and below corresponding NDVI initial thresholds
Interval range；
Unit 4.2：Optimal threshold extraction unit, for being iterated optimization respectively in the range of each experimental threshold values, is obtained
Altitudinal vegetation zone line of demarcation NDVI optimal thresholds；
In the present embodiment, optimal threshold extraction unit is mainly used in the range of research each experimental threshold values of area's NDVI images
The thought of maximum variance between clusters asks for corresponding altitudinal vegetation zone line of demarcation NDVI optimal thresholds respectively.
The basic thought of maximum variance between clusters is to split data into two classes using a certain threshold value, and selection makes side between two classes
Poor maximum threshold value is used as optimal threshold.When the area of two classes is more or less the same, effectively image can be split；But when two
When class area difference is larger, segmentation effect is then poor.
In the present embodiment, to avoid the disadvantages described above of maximum between-cluster variance, selection is used in the range of each experimental threshold values
The thought of maximum variance between clusters asks for corresponding optimal threshold respectively, can not only so reduce amount of calculation, also can guarantee that adjacent
The area of two classes is more or less the same, so as to more effectively split to image.In addition, initial in altitudinal vegetation zone line of demarcation NDVI
Optimal threshold extraction is carried out on the basis of threshold value, can also make final quantitative delineation result more accurate.
Unit 4.3：Reclassification unit, for carrying out Optimal-threshold segmentation to research area's NDVI images, and is weighed
Classification and assignment；
In the present embodiment, assignment scope is [0,1], and the different classes of of image is equidistantly assigned respectively from low to high according to height above sea level
It is worth for 1,0.6,0.3,0.
Unit 4.4：Probability graph generation unit, the neighborhood window for choosing particular size, is obtained by neighbor analysis method
Vegetation pattern probability graph；
In the present embodiment, the neighborhood window of particular size is the circular window that radius is 200m.
Neighborhood Statistics are that centered on grid to be calculated, the grid point value of certain limit around it (field) is counted
Analysis, is then output to the grid positions by end value and forms new raster map layer.The replacement of altitudinal vegetation zone mainly due to
Caused by the change of altitudinal gradient, grid is with respect to homogeneity in belt, with the rise of height above sea level, and heterogeneous grid gradually increases,
Until transition is another altitudinal vegetation zone type.Vegetation pattern probability graph progress Neighborhood Statistics are asked with the average value in neighborhood, then
Grid property value difference in same altitudinal vegetation zone is smaller, and grid property value should be close to two at different altitudinal vegetation zones
The average of altitudinal vegetation zone., then can not be by between different vegetation types when the field radius of use is too small compared with intermediate zone
Difference increase and protrude boundary line；And when the field radius of use is excessive, then this band vegetation pattern that some can be caused larger
Patch is smoothed processing, while also resulting in the reduction of Neighborhood Statistics figure resolution ratio.
In the present embodiment, by being tested to the different size of radius of neighbourhood, find when neighborhood window is that radius is
During 200m circular window, the pixel property value of same kind vegetation generally changes without too big, but some less non-
The patch of the zone of vegetation is smoothed, and the contrast between each altitudinal vegetation zone becomes apparent.
Unit 4.5：Boundary straight line extraction unit, the probability distribution threshold value for defining different vegetation types is extracted finally
Altitudinal vegetation zone line of demarcation.
In the present embodiment, unit 4.5 includes following subelement：
Subelement 4.5.1：Isopleth generates subelement, for based on the vegetation class acquired in 27.4 probability graph generation units
Type probability graph, generates the isopleth at intervals of 0.1；
Subelement 4.5.2：Initial line of demarcation obtains subelement, for based on the initial thresholds of altitudinal vegetation zone line of demarcation NDVI
Value, obtains the initial line of demarcation of research area's altitudinal vegetation zone；
Subelement 4.5.3：Optimize subelement at the beginning of line of demarcation, for by isopleth, the initial line of demarcation of altitudinal vegetation zone with grinding
Study carefully area's original remote sensing image figure to be overlapped, choose and the initial line of demarcation of altitudinal vegetation zone and remote sensing image visual interpretation vegetation class
Type matches optimal isopleth and is used as the altitudinal vegetation zone line of demarcation after optimization；
Subelement 4.5.4：Line of demarcation essence extracts subelement, the factor such as theoretical distribution height above sea level, slope aspect for combining vegetation,
Altitudinal vegetation zone line of demarcation after optimization is deleted, final altitudinal vegetation zone line of demarcation is obtained.
In the present embodiment, Neighborhood Statistics are split according only to vegetation distribution density to extract altitudinal vegetation zone point with isopleth
Boundary line, and according to ground investigation on the spot, research area's north slope (Schattenseite) evaporation capacity is small, moisture condition is good, and vegetation growing way is vigorous, Nan Po
For the less leeward slope of precipitation, the landscape of " northern woods south grass " is formed on the contrary.Therefore, vegetation is extracted according only to vegetation distribution density
Belt line of demarcation can have larger erroneous judgement.When quantitatively delineating research area progress altitudinal vegetation zone, also need to combine vegetation
The factor such as theoretical distribution height above sea level, slope aspect, the isopleth acquired in step 4.5.3 is judged and deleted, obtains final
Altitudinal vegetation zone line of demarcation.
Module 5：Output module, it is critical average with DEM for exporting altitudinal vegetation zone line of demarcation vector data and NDVI
Value.
Applicant is described in detail and described to embodiments herein with reference to Figure of description, but this area skill
Art personnel are it should be understood that above example is only the preferred embodiment of the application, and explanation is intended merely to help reader in detail
More fully understand that the present invention is spiritual, and the not limitation to the application protection domain, on the contrary, any based on the present application spirit
Any improvement or modification made should all fall within the protection domain of the application.
Claims (32)
1. a kind of altitudinal vegetation zone quantitatively delineates computational methods, for quantitatively delineating altitudinal vegetation zone distribution characteristics, its feature exists
In the computational methods comprise the following steps：
(1) remote sensing image data in altitudinal vegetation zone area to be extracted is obtained by satellite, passes through this area's digital elevation model
DEM extracts the corresponding gradient and slope aspect data；
(2) image procossing is carried out to the remote sensing image data, and carries out sample area screening；
(3) sample area scatter diagram is built, initial altitudinal vegetation zone cut off value is extracted；
(4) analyzed based on Neighborhood Statistics, extract altitudinal vegetation zone line of demarcation；
(5) output result image and data.
2. altitudinal vegetation zone according to claim 1 quantitatively delineates computational methods, it is characterised in that：
In the step (1), remote sensing image, digital complex demodulation, the gradient need to be vertical for same vegetation to be extracted with slope aspect
Band geographic coverage.
3. altitudinal vegetation zone according to claim 1 quantitatively delineates computational methods, it is characterised in that：
In the step (2), carrying out image procossing to the remote sensing image data includes following content：
2.1.1 radiation calibration, atmospheric correction and topographical correction pretreatment operation are carried out to remote sensing image；
2.1.2 vegetation-cover index NDVI is extracted based on pretreated remote sensing image；
2.1.3 visual fusion is carried out to NDVI, DEM, the gradient and slope aspect data.
4. altitudinal vegetation zone according to claim 3 quantitatively delineates computational methods, it is characterised in that：
In step 2.1.2, the calculation formula of the vegetation-cover index NDVI is：
In formula, NIR and R are respectively reflectance value of the vegetation near infrared band and infrared band.
5. altitudinal vegetation zone according to claim 3 quantitatively delineates computational methods, it is characterised in that：
In the step (2), sample area screening includes following content：
2.2.1 the fused data to NDVI, DEM, the gradient and slope aspect carries out gradient screening, and the screening gradient is not less than 5 degree of region
It is used as altitudinal vegetation zone to be extracted area；
2.2.2 slope aspect screening is carried out to altitudinal vegetation zone mountain area to be extracted, it is ensured that altitudinal vegetation zone area to be extracted is same slope
To.
6. altitudinal vegetation zone according to claim 1 quantitatively delineates computational methods, it is characterised in that：
The step (3) includes following content：
3.1 fused datas based on the altitudinal vegetation zone to be extracted area, using digital complex demodulation with from remote sensing image
The vegetation-cover index NDVI of middle extraction builds DEM-NDVI scatter diagrams；
3.2 pairs of DEM-NDVI scatter diagrams carry out density slice；
3.3, based on the DEM-NDVI scatter diagrams after the density slice, specific height above sea level are obtained using moving average method
NDVI average values and sliding average curve；
3.4 determine suitable matched curve function；
The variation tendency of 3.5 pairs of matched curves is analyzed, and finding out the separation of the different section of curve, to extract initial vegetation vertical
Straight band cut off value.
7. altitudinal vegetation zone according to claim 6 quantitatively delineates computational methods, it is characterised in that：
In the step 3.2, following content is included to DEM-NDVI scatter diagrams density slice：
3.2.1 it is scatterplot in the range of interval, statistical interval with 0.01 for certain specific height above sea level in DEM-NDVI scatter diagrams
Number；
3.2.2 the NDVI density values in every 0.01 interval are calculated, calculation formula is
In formula, Mi is the NDVI density values in i-th of NDVI interval, and Ni is the scatterplot number in i-th of NDVI interval,
Nmax is the maximum scatterplot number in all NDVI intervals；
3.2.3 aforesaid operations are carried out to all height above sea level in DEM-NDVI scatter diagrams, obtains DEM-NDVI density maps；
3.2.4 choose region of the NDVI density values more than 0.95 and be used as DEM-NDVI scatter diagram core spaces.
8. altitudinal vegetation zone according to claim 6 quantitatively delineates computational methods, it is characterised in that：
In the step 3.3, moving average method includes following content：
3.3.1 size is defined to be 100m window to ask for NDVI average values in the window ranges；
3.3.2 the window is entered into line slip by 5m sliding distance along elevation direction, so as to obtain equidistant height above sea level
NDVI average values；
3.3.3 all NDVI average points of above-mentioned acquisition are connected, you can obtain sliding average curve.
9. altitudinal vegetation zone according to claim 6 quantitatively delineates computational methods, it is characterised in that：
In the step 3.4, sliding average curve is fitted with n times Polynomical regressive equation；Matched curve and DEM-
The degree of fitting correlation coefficient r of NDVI sliding average curves2To weigh, N can determine that selection makes by multiple contrast experiment
r2Reach the minimum N values of maximum as suitable matched curve N values.
10. altitudinal vegetation zone according to claim 6 quantitatively delineates computational methods, it is characterised in that：
In the step 3.5, the separation of curve difference section is by asking flex point and second order to lead extreme point matched curve function
To be determined.
11. the altitudinal vegetation zone according to any claim in claim 6-9 quantitatively delineates computational methods, its feature exists
In：
In the step (4), based on Neighborhood Statistics analysis, extracting altitudinal vegetation zone line of demarcation includes following content：
4.1 obtain altitudinal vegetation zone line of demarcation NDVI initial thresholds and corresponding experimental threshold values scope；
4.2 are iterated optimization respectively in the range of each experimental threshold values, obtain altitudinal vegetation zone line of demarcation NDVI optimal thresholds；
4.3 pairs of sample area NDVI images are that the NDVI data in the range of sample area carry out Optimal-threshold segmentation, and are weighed
Classification and assignment, wherein, the Optimal-threshold segmentation refers to carry out image segmentation, institute to NDVI images based on NDVI optimal thresholds
State the different type that reclassification refers to represent NDVI image segmentation results with single value；
4.4 choose the neighborhood window of particular size, and vegetation pattern probability graph is obtained by neighbor analysis method；
4.5 define the probability distribution threshold value of different vegetation types, extract final altitudinal vegetation zone line of demarcation.
12. altitudinal vegetation zone according to claim 11 quantitatively delineates computational methods, it is characterised in that：
In the step 4.1, altitudinal vegetation zone line of demarcation NDVI initial thresholds are initial altitudinal vegetation zone matched curve not same district
NDVI values corresponding to section separation, experimental threshold values scope is the interval model of fluctuation 0.1 above and below corresponding NDVI initial thresholds
Enclose.
13. altitudinal vegetation zone according to claim 12 quantitatively delineates computational methods, it is characterised in that：
In the step 4.2, in the range of each experimental threshold values with maximum variance between clusters ask for respectively corresponding vegetation hang down
Straight band line of demarcation NDVI optimal thresholds.
14. altitudinal vegetation zone according to claim 13 quantitatively delineates computational methods, it is characterised in that：
Carrying out assignment to the NDVI images after Optimal-threshold segmentation in the step 4.3 includes following content：
4.3.1 the assignment scope to NDVI images after Optimal-threshold segmentation is [0,1]；
4.3.2 according to height above sea level 1 is equidistantly entered as respectively from low to high to the different classes of of NDVI images ..., 0；I.e. by vegetation cyclopentadienyl
Mi Qu is entered as 1, and no vegetation region is entered as 0, and middle transition class vegetation region is equidistantly entered as certain value in the range of [0,1].
15. altitudinal vegetation zone according to claim 14 quantitatively delineates computational methods, it is characterised in that：
The neighborhood window of particular size is the circular window that radius is 200m in the step 4.4.
16. altitudinal vegetation zone according to claim 15 quantitatively delineates computational methods, it is characterised in that:
Altitudinal vegetation zone line of demarcation is extracted in the step 4.5 includes following content：
4.5.1 vegetation pattern probability graph is based on, the isopleth at intervals of 0.1 is generated；
4.5.2 altitudinal vegetation zone line of demarcation NDVI initial thresholds are based on, the initial line of demarcation of research area's altitudinal vegetation zone is obtained；
4.5.3 isopleth, the initial line of demarcation of altitudinal vegetation zone and the regional original remote sensing image figure of altitudinal vegetation zone to be extracted are entered
Row superposition, chooses and optimal isopleth work is matched with the initial line of demarcation of altitudinal vegetation zone and remote sensing image visual interpretation vegetation pattern
For the altitudinal vegetation zone line of demarcation after optimization；
4.5.4 with reference to factors such as theoretical distribution height above sea level, the slope aspects of vegetation, the altitudinal vegetation zone line of demarcation after optimization is deleted
Subtract, obtain final altitudinal vegetation zone line of demarcation.
17. a kind of quantitative scoring system of altitudinal vegetation zone, for quantitatively delineating altitudinal vegetation zone distribution characteristics, the vegetation is vertical
The quantitative scoring system of band includes data read module, constituency module, scatterplot module, neighbor analysis module and output module, its
It is characterised by：
The data read module is used for remote sensing image, the digital complex demodulation for reading altitudinal vegetation zone area to be extracted, right
The remote sensing image read carries out image procossing, extracts vegetation-cover index, the gradient and the slope aspect number of remote sensing image after pretreatment
According to, and the fused data of vegetation-cover index NDVI, digital complex demodulation, the gradient and slope aspect is provided to constituency module；
The constituency module to being handled through data read module after remote sensing image carry out the gradient screening with slope aspect subregion so that
To altitudinal vegetation zone to be extracted area as research area, and the altitudinal vegetation zone mountain to be extracted after screening is provided to scatterplot module
Area's fused data；
The scatterplot module builds DEM- according to the altitudinal vegetation zone mountain area to be extracted fused data after the screening of constituency module
NDVI scatter diagrams, obtain altitudinal vegetation zone NDVI and DEM critical value, and provide altitudinal vegetation zone boundary to neighbor analysis module
Line NDVI initial thresholds；
The neighbor analysis module, for obtaining altitudinal vegetation zone line of demarcation, and it is vertical to the final vegetation of output module offer
Band line of demarcation vector data and NDVI and the critical average values of DEM；
The output module, for exporting altitudinal vegetation zone line of demarcation vector data and NDVI and the critical average values of DEM.
18. the quantitative scoring system of altitudinal vegetation zone according to claim 17, it is characterised in that：
In the data read module, remote sensing image, DEM, the gradient and slope aspect need to be same research area scope.
19. the quantitative scoring system of altitudinal vegetation zone according to claim 17, it is characterised in that：
The data read module includes image pre-processing unit, NDVI extraction units and visual fusion unit：
Wherein, described image pretreatment unit, it is pre- for carrying out radiation calibration, atmospheric correction and topographical correction to remote sensing image
Processing operation；
The NDVI extraction units are used for the vegetation-cover index NDVI for extracting remote sensing image after pretreatment；
The visual fusion unit is used to carry out visual fusion to NDVI, DEM, the gradient and slope aspect data.
20. the quantitative scoring system of altitudinal vegetation zone according to claim 19, it is characterised in that：
The NDVI is that vegetation index is vegetation-cover index, and calculation formula is：
In formula, NIR and R are respectively reflectance value of the vegetation near infrared band and infrared band.
21. the quantitative scoring system of altitudinal vegetation zone according to claim 17, it is characterised in that：
The constituency module includes gradient screening unit and slope aspect screening unit：
The gradient screening unit is used to carry out gradient screening to the fused data of NDVI, DEM, the gradient and slope aspect, screens the gradient
Region not less than 5 degree is used as altitudinal vegetation zone mountain area to be extracted；
The slope aspect screening unit, for carrying out slope aspect screening to altitudinal vegetation zone mountain area to be extracted, it is ensured that vegetation to be extracted is hung down
Straight band mountain area is same slope aspect.
22. the quantitative scoring system of altitudinal vegetation zone according to claim 17, it is characterised in that：
The scatterplot module includes scatter diagram generation unit, density slice unit, moving average unit, matched curve list
Member and separation extraction unit：
Wherein, melt in the altitudinal vegetation zone mountain area to be extracted that the scatter diagram generation unit is used to read after the module screening of constituency
Data are closed, and DEM-NDVI scatter diagrams are built using DEM and NDVI；
The density slice unit is used to carry out density slice to the DEM-NDVI scatter diagrams；
The moving average unit is used to carry out moving average to the DEM-NDVI scatter diagrams after the density slice, and obtains
Take the NDVI average values and sliding average curve of specific height above sea level；
The matched curve unit is used to determine suitable matched curve function；
The separation extraction unit is used to analyze the variation tendency of matched curve, finds out the boundary of the different sections of curve
Point.
23. the quantitative scoring system of altitudinal vegetation zone according to claim 22, it is characterised in that：
The density slice unit includes scatterplot statistics subelement, density computation subunit, density map subelement and core space life
Into subelement；
Wherein, the scatterplot statistics subelement is used for the scatterplot for reading the specific height above sea level of certain in DEM-NDVI scatter diagrams, is with 0.01
Interval, counts the scatterplot number in certain specific height above sea level interval；
The density computation subunit is used to calculate the NDVI density values in 0.01 interval, and calculation formula is
In formula, Mi is the NDVI density values in i-th of NDVI interval, and Ni is the scatterplot number in i-th of NDVI interval,
Nmax is the maximum scatterplot number in all NDVI intervals；
The density map subelement is used to carry out aforesaid operations to all height above sea level in DEM-NDVI scatter diagrams, obtains DEM-NDVI
Density map；
The core space generation subelement is used to choose region of the NDVI density values more than 0.95 as the distribution of DEM-NDVI scatterplots
Kernel of graph heart district.
24. the quantitative scoring system of altitudinal vegetation zone according to claim 23, it is characterised in that：
It is single that the moving average unit includes window mean value computation subelement, equidistant mean value computation subelement and curve generation
Member；
Wherein, the window mean value computation subelement is used to define size for 100m window to ask for NDVI in the window ranges
Average value；
The equidistant mean value computation subelement is used to the window entering line slip along elevation direction by 5m sliding distance, obtains
The NDVI average values of equidistant height above sea level；
The curve generates are used to all NDVI average points of above-mentioned acquisition being connected, and obtain sliding average curve.
25. the quantitative scoring system of altitudinal vegetation zone according to claim 22, it is characterised in that：
In matched curve unit, sliding average curve is carried out curve fitting using n times Polynomical regressive equation, and will be intended
Close the degree of fitting correlation coefficient r of curve and DEM-NDVI sliding average curves2Weighed, wherein, N is by repeatedly contrasting
Test to determine, selection makes r2Reach the minimum N values of maximum as suitable matched curve N values.
26. the quantitative scoring system of altitudinal vegetation zone according to claim 22 or 26, it is characterised in that：
In separation extraction unit, determine matched curve not by asking flex point and second order to lead extreme point matched curve function
With the separation of section.
27. the quantitative scoring system of altitudinal vegetation zone according to claim 17 or 22, it is characterised in that：
The neighbor analysis module includes threshold range acquiring unit, optimal threshold extraction unit, reclassification unit, probability graph life
Into unit and boundary straight line extraction unit；
Wherein, the threshold range acquiring unit is used to obtain altitudinal vegetation zone line of demarcation NDVI initial thresholds and corresponding experiment
Threshold range；
The optimal threshold extraction unit is used to be iterated optimization respectively in the range of each experimental threshold values, obtains altitudinal vegetation zone
Line of demarcation NDVI optimal thresholds；
The reclassification unit is used to carry out Optimal-threshold segmentation to the NDVI data in the range of sample area, and is divided again
Class and assignment；
The probability graph generation unit is used for the neighborhood window for choosing particular size, and it is general to obtain vegetation pattern by neighbor analysis method
Rate figure；
The boundary straight line extraction unit, the probability distribution threshold value for defining different vegetation types extracts final vegetation vertical
Band line of demarcation.
28. the quantitative scoring system of altitudinal vegetation zone according to claim 27, it is characterised in that：
In threshold range acquiring unit, by matched curve difference section determined by separation extraction unit in scatterplot module
NDVI values corresponding to separation are as altitudinal vegetation zone line of demarcation NDVI initial thresholds, above and below corresponding NDVI initial thresholds
The interval range of fluctuation 0.1 is used as experimental threshold values scope.
29. the quantitative scoring system of altitudinal vegetation zone according to claim 27 or 28, it is characterised in that：
The optimal threshold extraction unit includes inter-class variance subelement, and the inter-class variance subelement is used in sample area scope
Corresponding altitudinal vegetation zone is asked for respectively with the thought of maximum variance between clusters in the range of interior each experimental threshold values of NDVI images
Line of demarcation NDVI optimal thresholds.
30. the quantitative scoring system of altitudinal vegetation zone according to claim 27, it is characterised in that：
The reclassification unit is used to determine assignment scope and the sorted image of counterweight carries out assignment.
31. the quantitative scoring system of altitudinal vegetation zone according to claim 27, it is characterised in that：
In the probability graph generation unit, the neighborhood window of particular size is the circular window that radius is 200m.
32. the quantitative scoring system of altitudinal vegetation zone according to claim 27, it is characterised in that：
The boundary straight line extraction unit includes isopleth generation subelement, initial line of demarcation and optimized at the beginning of obtaining subelement, line of demarcation
Optimize subelement at the beginning of subelement, line of demarcation and line of demarcation essence extracts subelement；
Wherein, isopleth generation subelement is used for based on the vegetation pattern probability graph acquired in probability graph generation unit, between generation
It is divided into 0.1 isopleth；
The initial line of demarcation, which obtains subelement, to be used to be based on altitudinal vegetation zone line of demarcation NDVI initial thresholds, is obtained research area and is planted
By the initial line of demarcation of belt；
Optimizing subelement at the beginning of the line of demarcation is used for isopleth, the initial line of demarcation of altitudinal vegetation zone and the research original remote sensing shadow in area
As figure is overlapped, choose matched with the initial line of demarcation of altitudinal vegetation zone and remote sensing image visual interpretation vegetation pattern most preferably etc.
Value line is used as the altitudinal vegetation zone line of demarcation after optimization；
The line of demarcation essence, which extracts subelement, to be used to combine the factors such as theoretical distribution height above sea level, the slope aspect of vegetation, to the plant after optimization
Deleted by belt line of demarcation, obtain final altitudinal vegetation zone line of demarcation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710467468.5A CN107330898B (en) | 2017-06-20 | 2017-06-20 | Quantitative marking calculation method and system for vegetation vertical zone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710467468.5A CN107330898B (en) | 2017-06-20 | 2017-06-20 | Quantitative marking calculation method and system for vegetation vertical zone |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107330898A true CN107330898A (en) | 2017-11-07 |
| CN107330898B CN107330898B (en) | 2020-04-28 |
Family
ID=60194425
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710467468.5A Active CN107330898B (en) | 2017-06-20 | 2017-06-20 | Quantitative marking calculation method and system for vegetation vertical zone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107330898B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111861836A (en) * | 2020-07-20 | 2020-10-30 | 云南财经大学 | Three-dimensional mountain land planning method and device, storage medium and computer equipment |
| CN112347978A (en) * | 2020-11-25 | 2021-02-09 | 中国科学院东北地理与农业生态研究所 | Ecological system attribute component composition structure description method based on remote sensing vegetation index |
| CN112907587A (en) * | 2021-04-01 | 2021-06-04 | 西南石油大学 | High mountain forest line extraction method based on Otsu and edge detection algorithm of GEE |
| CN113433553A (en) * | 2021-06-23 | 2021-09-24 | 哈尔滨工程大学 | Precise navigation method for multi-source acoustic information fusion of underwater robot |
| CN115019184A (en) * | 2022-07-28 | 2022-09-06 | 北京卫星信息工程研究所 | Remote-sensing-image-based stony desertification degree automatic grading method and device |
| WO2023087971A1 (en) * | 2021-11-18 | 2023-05-25 | 东南大学 | Contour-line matching method based on sliding window data backtracking |
| CN116468321A (en) * | 2023-04-23 | 2023-07-21 | 河北省科学院地理科学研究所 | Topography comprehensive index method for representing vegetation growth distribution rule |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105957115B (en) * | 2016-05-18 | 2018-06-08 | 河北工程大学 | Main crops production Remotely sensed acquisition method under broad sense DEM thoughts |
| CN106384116A (en) * | 2016-08-29 | 2017-02-08 | 北京农业信息技术研究中心 | Terahertz imaging based plant vein recognition method and device |
-
2017
- 2017-06-20 CN CN201710467468.5A patent/CN107330898B/en active Active
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111861836A (en) * | 2020-07-20 | 2020-10-30 | 云南财经大学 | Three-dimensional mountain land planning method and device, storage medium and computer equipment |
| CN111861836B (en) * | 2020-07-20 | 2022-10-18 | 云南财经大学 | Three-dimensional mountain land planning method and device, storage medium and computer equipment |
| CN112347978A (en) * | 2020-11-25 | 2021-02-09 | 中国科学院东北地理与农业生态研究所 | Ecological system attribute component composition structure description method based on remote sensing vegetation index |
| CN112347978B (en) * | 2020-11-25 | 2022-06-03 | 中国科学院东北地理与农业生态研究所 | Ecological system attribute component composition structure description method based on remote sensing vegetation index |
| CN112907587A (en) * | 2021-04-01 | 2021-06-04 | 西南石油大学 | High mountain forest line extraction method based on Otsu and edge detection algorithm of GEE |
| CN112907587B (en) * | 2021-04-01 | 2022-03-01 | 西南石油大学 | High mountain forest line extraction method based on Otsu and edge detection algorithm of GEE |
| CN113433553A (en) * | 2021-06-23 | 2021-09-24 | 哈尔滨工程大学 | Precise navigation method for multi-source acoustic information fusion of underwater robot |
| CN113433553B (en) * | 2021-06-23 | 2022-08-02 | 哈尔滨工程大学 | Precise navigation method for multi-source acoustic information fusion of underwater robot |
| WO2023087971A1 (en) * | 2021-11-18 | 2023-05-25 | 东南大学 | Contour-line matching method based on sliding window data backtracking |
| US11835344B2 (en) | 2021-11-18 | 2023-12-05 | Southeast University | Contour line matching method based on sliding window data backtracking |
| CN115019184A (en) * | 2022-07-28 | 2022-09-06 | 北京卫星信息工程研究所 | Remote-sensing-image-based stony desertification degree automatic grading method and device |
| CN116468321A (en) * | 2023-04-23 | 2023-07-21 | 河北省科学院地理科学研究所 | Topography comprehensive index method for representing vegetation growth distribution rule |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107330898B (en) | 2020-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107330898A (en) | Altitudinal vegetation zone quantitatively delineates computational methods and system | |
| CN109101894B (en) | A kind of remote sensing image clouds shadow detection method that ground surface type data are supported | |
| CN104751478B (en) | Object-oriented building change detection method based on multi-feature fusion | |
| CN110378909B (en) | Single wood segmentation method for laser point cloud based on Faster R-CNN | |
| CN109948596A (en) | A method of rice identification and crop coverage measurement are carried out based on vegetation index model | |
| CN105184423B (en) | A kind of wind power plant cluster wind speed forecasting method | |
| CN104463164B (en) | It is a kind of based on umbrella frame method and crown height than trees canopy structure information extracting method | |
| CN106650689A (en) | Coastal city time sequence land utilization information extracting method | |
| CN109583516A (en) | A kind of space and time continuous PM2.5 inversion method based on ground and moonscope | |
| CN111091079B (en) | TLS-based method for measuring vegetation advantage single plant structural parameters in friable region | |
| CN108896021B (en) | Method for extracting artificial forest stand structure parameters based on aerial photogrammetry point cloud | |
| CN111340826A (en) | Single tree crown segmentation algorithm for aerial image based on superpixels and topological features | |
| CN108764012A (en) | The urban road shaft recognizer of mobile lidar data based on multi-frame joint | |
| CN107330413A (en) | A kind of mother drug plants recognition methods based on remote sensing technology | |
| CN112861810B (en) | Artificial forest planting time automatic detection method based on time sequence remote sensing observation data | |
| CN110988909A (en) | TLS-based vegetation coverage determination method for sandy land vegetation in alpine and fragile areas | |
| CN109856021A (en) | A kind of PM2.5 inversion method and monitoring region segmentation method | |
| CN109961470A (en) | A kind of accurate estimating and measuring method of live standing tree leaf attribute based on laser point cloud | |
| CN104573662B (en) | A kind of cloud sentences method and system | |
| CN102592118B (en) | Automatic detection method for time emergence of seedling of corns | |
| CN109242786A (en) | A kind of automation morphologic filtering method suitable for urban area | |
| CN110598707B (en) | Single plant tree crown segmentation method for airborne laser point cloud water pouring spreading and energy function control | |
| CN112380994A (en) | Artificial forest spatial pattern recognition method based on time sequence classification and spatial analysis | |
| Zhu et al. | Research on deep learning individual tree segmentation method coupling RetinaNet and point cloud clustering | |
| CN106023178A (en) | Method for detecting single tree in remote sensing data based on gradient direction clustering |
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 |