CN104615977B - The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology - Google Patents
The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology Download PDFInfo
- Publication number
- CN104615977B CN104615977B CN201510037425.4A CN201510037425A CN104615977B CN 104615977 B CN104615977 B CN 104615977B CN 201510037425 A CN201510037425 A CN 201510037425A CN 104615977 B CN104615977 B CN 104615977B
- Authority
- CN
- China
- Prior art keywords
- winter wheat
- remote sensing
- pixel
- resolution
- modis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to remote sensing monitoring technical field, and in particular to a kind of winter wheat remote sensing recognition method for comprehensively utilizing crucial Aspection character and fuzzy classification technology.This method includes：Data prediction, the abundance figure under preparation research area coarse resolution, pixel is obtained under the middle high-resolution yardstick in research area for steps such as the degree of membership of winter wheat, Comprehensive Evaluations.The present invention combines the method based on the aspect rhythm and pace of moving things（Utilize the jump of low resolution remote sensing）And fuzzy classification technology（Utilize the spectral information of middle high-resolution remote sensing）Obtain the middle high-resolution recognition result with definite spatial distribution, the shortcomings that compensate for two methods each, both solved fuzzy classification technology pixel belong to all kinds of probability it is suitable when uncertain problem, the abundance figure for solving the problems, such as to obtain using Aspection character again can not show the definite spatial distribution of crop, and new monitoring, evaluation measures are provided for the remote sensing monitoring of winter wheat.
Description
Technical field
The invention belongs to remote sensing monitoring technical field, and in particular to one kind comprehensively utilizes crucial Aspection character and fuzzy classification
The winter wheat remote sensing recognition method of technology.
Background technology
In the prior art, the method based on remote sensing technology identification agrotype mainly has method and base based on spectral information
In the method for the aspect rhythm and pace of moving things.
Method for distinguishing is known based on spectral information and is mainly used in Moderate-High Spatial Resolution Remote Sensing Image, principle is the system using pixel value
Count feature and carry out Classification and Identification, but due to the presence of " the different spectrum of jljl " and " foreign matter is with spectrum " phenomenon, cause recognition result to occur inclined
Difference.
Recognition methods based on aspect prosodic feature is mainly used in being formed the low spatial resolution remote sensing shadow of time series
Picture, principle is the difference with other vegetation growth rhythm and pace of moving things using crop, but image resolution is low to cause recognition result due to existing
The defects of precision is not high, thus often utilize the recognition methods of sub- pixel, such as linear spectral unmixing approach, based on the crucial phenological period
Area index method etc..But the result obtained using these methods is all an abundance figure, i.e., makees object plane in coarse resolution pixel
Long-pending ratio, and the specific distribution of crop can not be definitely determined, thus specifically used go up still has larger inconvenience.
Because two class methods respectively have advantage and disadvantage, thus it also will often be based on spectral information method and be based on aspect rhythm and pace of moving things method knot
Close and use.But the combination of two methods at present is still mainly limited to provide mixed pixel using middle high-resolution image
The pure unit decomposited, then carry out Decomposition of Mixed Pixels using the spectral characteristic of low-resolution data.Although this combination tool
There is the advantages of certain, but do not give full play to the advantage of two methods yet, and its recognition result is only the cultivated area of crop,
Crop spatial distribution truly is not extracted, thus is actually used still relatively limited.
Fuzzy classification technology is identified applied to agrotype, the uncertain problem belonged to for pixel in remote sensing image provides
Useful theoretical foundation, its principle are：When a pixel can belong to different classes simultaneously, key is to determine the picture
Pixel is finally determined as that maximum one kind of degree of membership by member for all kinds of degrees of membership.But the problem of this method, is, when
When the class of highest two or substantially suitable multiclass degree of membership be present, also deposited even if the pixel is belonged into that maximum one kind of degree of membership
Obvious uncertain.
In the prior art, it is a kind of more quick and directly perceived the cultivated area of winter wheat to be monitored using remote sensing image
Technological means, and then there are multiple technologies means in the deciphering to remote sensing image, but because all kinds of technological means itself are intrinsic
Advantage and disadvantage, thus often need one or more kinds of technical tie-ups using and while carry out mutually correction and could obtain accurate monitor
Data, and in the prior art, there is not yet comprehensive crucial Aspection character and fuzzy classification technology are used for winter wheat remote sensing recognition
Relevant report.
The content of the invention
Present invention aims at provide a kind of winter wheat remote sensing for comprehensively utilizing crucial Aspection character and fuzzy classification technology
Recognition methods.The present invention major technique thinking be：Obtained using crucial Aspection character rich under research area's coarse resolution yardstick
Degree figure；Using fuzzy classification technology obtain study area's middle high-resolution yardstick under pixel for winter wheat degree of membership；In synthesis
Two kinds of results are stated, " foreign matter is with spectrum " phenomenon is corrected in fuzzy classification using Aspection character information, utilizes pixel in fuzzy classification
Degree of membership determines the particular location of abundance figure neutron pixel, by the mutual supplement with each other's advantages of two methods, so as to obtain middle high-resolution chi
The definite spatial distribution of the lower research area winter wheat of degree.
The technical solution used in the present invention is specifically described as follows.
The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology, comprises the following steps：
（1）Data prediction
By TM middle high-resolutions image, the data in projection and coordinate system are completely registering with MODIS coarse resolutions image；
Specifically such as：TM data resamplings are set to 25 meters, matched somebody with somebody completely with 250 meters of data of MODIS in projection and coordinate system
It is accurate so that each MODIS pixel can correspond to 10 × 10 TM pixels on locus；
（2）Abundance figure under preparation research area coarse resolution
A. according to research area winter wheat plantation situation, in the case where ensuring precision conditions, typical sample investigation MODIS is laid in field
Winter wheat planting area ratio in one pixel of coarse resolution yardstick, sample is provided for regression analysis；High-resolution can also be used
Remote sensing image is interpreted to winter wheat plantation situation, makes the pixel percentage of MODIS coarse resolution yardsticks based on this
Data are as sample, so as to replace field operation to lay sample prescription；
B. extraction reflects the period in winter wheat typical case's phenological period from MODIS NDVI time serieses, utilizes the period
MODIS NDVI data establish typical phenological period slope image；The winter wheat typical case phenological period be preferably winter wheat period of seedling establishment extremely
Winter wheat boot stage；
C. regression model is established using typical phenological period slope image obtained by step B and sample data obtained by step A, and will
It expands to whole research area, so as to obtain studying the abundance figure of the winter wheat planting area in area；
（3）Degree of membership of the pixel for winter wheat under the middle high-resolution yardstick in acquisition research area
A. according to TM data, using fuzzy classification technology, the degree of membership based on bayesian criterion determines method, obtains each picture
Member is under the jurisdiction of the degree of membership of winter wheat；Circular is as follows：
Assuming that vector in classification image, whereinX TFor vectorXTransposition, research is divided intom
Class（k i , i=1,2 ..., m）, according to Bayesian formula,XBelonged under conditions of appearancek i The ownership probability of class is：
,
VectorXIn classificationk i Conditional probability density function be：
,
In formula,P(k i ) it is classificationk i Prior probability；P(X/k i ) it is classificationk i It is middle vector occurXProbability；N is characterized
The dimension in space；mTo study area's classification number；μ i For classificationk i Training sample mean vector；For classificationk i Train sample
This covariance matrixDeterminant；
B. Data correction, " the different spectrum of jljl " phenomenon is eliminated；For " the different spectrum of jljl " phenomenon, for example, winter wheat due to weather and
The conditions such as growing way difference and there is the difference of spectrum, can by increasing investigation sample, respectively calculate degree of membership method add
To solve, so that it is guaranteed that all pixels for belonging to winter wheat can extract；
（4）Comprehensive Evaluation, i.e. combining step（2）Abundance figure and step（3）It is subordinate to degrees of data to winter wheat kind in remote sensing figure
Plant situation is judged
A. in each MODIS pixels, the TM pixels corresponding to it are sorted from high to low to the degree of membership of winter wheat；
, then will be under the conditions of middle high-resolution, on locus if B. the winter wheat abundance of a MODIS pixel is F%
Corresponding 10 × 10, the preceding F% pixel to be sorted by step A winter wheat degree of membership be determined as winter wheat, F values are according to step
Suddenly（2）In abundance figure determine；
C. whole research area is expanded in this way, you can to obtain the cultivated area of winter wheat and distribution.
The winter wheat remote sensing recognition method of the crucial Aspection character of synthesis provided by the present invention and fuzzy classification technology, it is comprehensive
Method based on the aspect rhythm and pace of moving things（Utilize the jump of low resolution remote sensing）And fuzzy classification technology（Utilize middle high-resolution
The spectral information of remote sensing）, the middle high-resolution recognition result with definite spatial distribution is obtained, it is respective to compensate for two methods
Shortcoming, both solved fuzzy classification technology pixel belong to all kinds of probability it is suitable when uncertain problem, solve profit again
The abundance figure obtained with Aspection character can not show the definite spatial distribution problem of crop, be provided for the remote sensing monitoring of winter wheat
New monitoring, evaluation measures, while new reference also is provided for the remote sensing monitoring of other agrotypes, and it is used
The remote sensing image data of middle low resolution yardstick is easily obtained, and is easy to carry out actual monitoring application on certain regional scale,
Thus there is preferable application value.
Brief description of the drawings
Fig. 1 is the general technological system figure of the present invention；
Fig. 2 lays situation for the field sample prescription of Foundation of Luoyang；
Fig. 3 is 2009-2010 years winter wheat growth seasons of Foundation of Luoyang different vegetation types NDVI time-serial positions, from
In can reflect the Aspection characters of different vegetation；
Fig. 4 is that the winter wheat of comprehensive crucial Aspection character and fuzzy classification technology identifies schematic diagram；Wherein Fig. 4（A）It is one
Individual one pixel of coarse resolution image and intermediate-resolution Pixel domain corresponding relation, entirely represent 250 meters of pixels of a MODIS,
Each lattice represents a pixel of 25 meters of resolution ratio, each 25 meters of resolutions that digitized representation is obtained using fuzzy technology
Rate pixel is under the jurisdiction of the degree of membership of winter wheat；Fig. 4（B）Recognition result during MODIS pixel abundance F%=58% is represented, grey is dashed forward
Go out the particular location of the winter wheat component identified for this MODIS pixel of display；
Fig. 5 is the metrical scale winter wheat abundance result of calculations of MODIS 250；
Fig. 6 is degree of membership result of calculation of the metrical scale pixels of TM 25 to winter wheat；
Fig. 7 is that the Growing season winter wheat of 2009-2010 Foundation of Luoyang remote sensing figures is identified using the inventive method
As a result.
Embodiment
With reference to embodiment the present invention will be further explained explanation.
Before specific embodiment is introduced, the concept brief explanation of the application Satellite image resolution is described as follows：Pin
It is low for different satellites to satellite image resolution ratio（Slightly）Resolution ratio, high-resolution implication are not fully consistent, with regard to the application
For, it is considered that the image of this hundred meter levels resolution ratio of MODIS is low resolution；This ten meter levels images of TM are intermediate-resolution,
Meter level resolution image is high-resolution.Thus during non-specified otherwise, the application is i.e. as standard.
Embodiment
The present embodiment is using Foundation of Luoyang as research area, using the present invention to the related distant of the region 2009-2010
Sense image has carried out interpreting and having carried out checking analysis, is briefly discussed below.
The major technique thinking of the present invention is as shown in Figure 1：Resampling, such as TM points are carried out according to TM remote sensing images first
Resolution is set as 25 meters, obtains TM25 rice image datas, and this data then is carried out into space with MODIS coarse resolutions image matches somebody with somebody
Standard, while degree of membership calculating is carried out to TM25 rice image data.Second, all kinds of vegetation sequential are obtained according to MODIS time serieses
Curve, winter wheat critical period is selected, obtain the slope image of winter wheat critical period；Carried out simultaneously according to MODIS images outer
Remote sensing image is sampled after industry sample prescription is laid investigation or interpreted according to high-resolution, obtains the winter wheat under MODIS yardsticks
Area ratio；Returned with reference to the winter wheat area ratio data that the slope image of winter wheat critical period obtains with sample investigation
Return analysis, obtain the identification model of winter wheat, further obtain winter wheat abundance figure.3rd, entered according to TM25 rice image datas
Row degree of membership calculates, and further belongs to probability image.Finally, the degree of membership result and MODIS chis of comprehensive TM25 rice image datas
Abundance figure result under degree carries out judging the recognition result for obtaining winter wheat, and judgment criteria is：Corresponding to each MODIS pixel
10 × 10 TM25 rice pixels, are arranged from big to small by degree of membership, according to the Abundances F% of the MODIS pixels, 10 × 10 by before
× F% pixel is determined as winter wheat.
The present invention is specifically described as follows.
Comprehensive crucial Aspection character and the method for fuzzy classification technology remote sensing recognition winter wheat, comprise the following steps：
（1）Remotely-sensed data pre-processes
TM data resamplings are set to 25 meters, and it is completely registering with 250 meters of data of MODIS in projection and coordinate system, make
10 × 10 TM pixels can be corresponded on locus by obtaining each MODIS pixel.
（2）Abundance figure under preparation research area coarse resolution
A. according to research area winter wheat plantation situation, in the case where ensuring precision conditions, typical sample investigation MODIS is laid in field
Winter wheat planting area ratio in one pixel of yardstick, sample is provided for regression analysis；To reduce field investigation, height can also be used
Resolution remote sense image is interpreted to winter wheat plantation situation, makes the pixel percentage number of MODIS yardsticks based on this
According to as sample, so as to replace field operation to lay sample prescription.
For specific the present embodiment, inventor has carried out field investigation for the winter wheat plantation situation of Foundation of Luoyang, has
The distribution of body sample prescription sets as shown in Figure 2.
Sample prescription sets 55 altogether, wherein size is 250 meters × 250 meters to each sample prescription on the spot, while sample prescription is selected, set
Different size planting proportion is ensured in journey（Area ratio in sample prescription shared by winter wheat）Sample prescription have laying.Actual sample prescription cloth
If during, the sample prescription that a general county lays 20 or so can meet to require.
B. extraction reflects the period in winter wheat typical case's phenological period from MODIS NDVI time serieses, utilizes the period
MODIS NDVI data establish typical phenological period slope image.
Specifically, the NDVI time sequences based on the MODIS NDVI time series datas extraction pure pixel of different vegetation types
Row curve, as shown in Figure 3.
Due to studying in area's same period in the NDVI time-serial positions of different vegetation types, the curve of winter wheat
Other vegetation patterns are differed markedly from, can clearly distinguish the characteristic point of A, B, C, D, E, F main points.If thus
One MODIS pixel is closer to the pure pixel of winter wheat, then its NDVI curve and standard curve（During the NDVI of the pure pixel of winter wheat
Between sequence curve）It is more similar；But with increasing for other vegetation patterns has been mixed into pixel, then it can cause song between key feature points
The change of line slope, this identifies winter wheat theoretical foundation also with aspect prosodic feature.
Require emphasis and illustrate, mostly in existing research is to identify winter wheat by comprehensive AB sections and EF sections, i.e.,：Such as
Other kinds of atural object is mixed into fruit pixel can all cause the change of AB sections and EF slope over 10, be built eventually through with measured data
Winter wheat is identified in regression model.But more gross error during actual monitoring often be present, main reason is that AB sections
It is very big with the difference of other atural objects NDVI in EF sections, when causing the different type atural object that same ratio is mixed into pixel, pixel
Value can also be very different, so that the change of AB sections and EF slope over 10 is different, that is to say, that when being mixed into it in pixel
The timing of ratio one of his atural object, it should make AB sections and EF slope over 10 that corresponding change occur in theory, crop could be identified very well,
But actually because mixed type of ground objects is different AB sections can be caused different with EF slope over 10, so as to cause recognition result to exist
Error.In brief, identify that winter wheat easily has larger error with AB sections and EF sections.
It has been recognised by the inventors that in Fig. 3 C time points be generally corresponding to the period of seedling establishment of winter wheat, D time points are generally corresponding to winter wheat
Boot stage, and the starting point of this time point to be other vegetation patterns start growth, thus when in CD sections being winter wheat fast-growth
Phase, and the NDVI values of other atural objects are relatively stable, and it is not clearly, so that being mixed into other vegetation patterns of pixel to differ
Ratio it is more, the slope of CD sections will be smaller, and the influence of other vegetation pattern differences is preferably minimized.Thus invent
People thinks can preferably be solved using CD sections structure identification model commonly used in the prior art small using AB sections and EF sections judgement winter
The problem of larger error during wheat be present.
By calculating, the present invention utilizes the regression model that the slope of CD sections is established with the pixel percentage of the winter wheat of actual measurement
It is as follows：
；
Wherein,FWWFor winter wheat pixel area percentage；NDVI D-NDVI CFor the slope of CD sections（In general, CD when
Between section be the unit time）；A and b is coefficient to be asked.
Further, the MODIS NDVI data structure in research on utilization area represents the image data of CD slope over 10, and with open country
Outer measured data carries out recurrence calculating, obtains coefficient a and b to be asked, and model extension can be obtained into research to whole research area
The winter wheat planting area abundance figure of area's MODIS yardsticks.In general, if research area's area is very big, ecology point can be utilized
Area builds different models and is identified.
For specific the present embodiment, established using the MODIS NDVI images for reflecting CD slope over 10 and 55 sample prescription data
Regression model is as follows：
；
Further carry out returning checking analysis, carry out coefficient of determination（Square of coefficient correlation）Calculate, judge linear regression
Fitting degree（The purpose is to for illustrate with independent variable explain dependent variable make a variation degree, R2Bigger fitting degree is higher）, R2
=0.8, illustrate that there is good fitting effect.
This regression model is expanded into whole research area, you can obtain studying the abundance figure of the winter wheat planting area in area,
As shown in Figure 5.
（3）Degree of membership of the pixel for winter wheat under the middle high-resolution yardstick in research area
A. according to the 25 of TM resamplings meters of resolution images, classified using fuzzy classification technology, obtain each pixel and be subordinate to
Belong to the degree of membership of winter wheat.
For specific the present embodiment, using fuzzy classification technology, the degree of membership based on bayesian criterion determines method, finally
Degree of membership image of all categories is obtained, here only with the degree of membership image for being under the jurisdiction of " winter wheat ".Computational methods are as follows：
Assuming that vector in classification image, research is divided intomClass（k i , i=1,2 ..., m）, according to Bayesian formula,XBelonged under conditions of appearancek i The ownership probability of class（XFork i The degree of membership of class）For：
,
VectorXIn classificationk i Conditional probability density function be：
,
In formula,P(k i ) it is classificationk i Prior probability；P(X/k i ) it is classificationk i It is middle vector occurXProbability；N is characterized
The dimension in space；mTo study area's classification number；μ i For classificationk i Training sample mean vector；For classificationk i Train sample
This covariance matrixDeterminant.
Using this method, on the basis of resampling is the TM images of 25 meters of resolution ratio calculating each pixel belongs to winter wheat
Ownership probability, as a result as shown in Figure 6.
B. for " the different spectrum of jljl " phenomenon, as there is spectrum due to the difference of weather and growing way condition in winter wheat
It difference, can be solved by multiselect sample, ensure that all pixels for belonging to winter wheat can extract.
（4）Combining step（2）Abundance figure and step（3）It is subordinate to degrees of data to comment winter wheat plantation situation in remote sensing figure
Sentence
A. in each MODIS pixels, will corresponding to 10 × 10 TM pixels according to its degree of membership to winter wheat from height
To low sequence.
For specific the present embodiment, such as Fig. 4（A）It show 250 meters of pixels of a MODIS and 25 meters of TM after resampling
The spatial correspondence of pixel, the numeral in lattice are that the TM pixels obtained based on fuzzy classification technology are under the jurisdiction of winter wheat
Degree of membership.
, then will be under the conditions of middle high-resolution, institute on locus if B. the winter wheat abundance of the MODIS pixels is F%
Corresponding 10 × 10, by step A winter wheat degree of membership sort preceding F% pixel be determined as winter wheat, i.e., by corresponding to it
Preceding 100 × F% the pixel of TM yardsticks is determined as winter wheat, this method is expanded into whole research area, you can to obtain winter wheat
Cultivated area and distribution.
Specifically, if the value of the MODIS pixels in abundance figure is 58%, then winter wheat in the MODIS pixels is shown
Area ratio be 58%, and 25 meters of pixel quantity of the TM corresponding to the pixel are 100, then corresponding TM yardsticks winter wheat picture
First number should be 100 × 58%=58, and this 100 TM pixels are arranged from big to small by degree of membership, take preceding 58 pixels to determine
For winter wheat pixel, as a result such as Fig. 4（B）Shown, the part that wherein grey highlights is the TM pixels for being identified as winter wheat.
Above-mentioned processing is done to all pixels in whole research area, you can so that the winter wheat for studying area to be identified.Root
According to this method, it is as shown in Figure 7 that research area's winter wheat recognition result is calculated.
（5）Checking
Crucial Aspection character and the side of fuzzy classification technology remote sensing recognition winter wheat are utilized for checking is provided by the present invention
The accuracy of method, inventor further weigh recognition result in terms of the positional precision and area precision two, are briefly discussed below.
What positional precision was reflected is to be identified as the levels of precision of the pixel of winter wheat in position using this method.Checking
Method is：Some checking sampling points are generated at random in research area, and using the result of visual interpretation as true value, positional precision is defined as
The ratio of the number of samples correctly identified on locus and total number of samples, calculation formula are as follows：
；
In formula,ApFor positional precision；NUM R To be correctly identified as the number of samples of winter wheat；NUM T For total number of samples.
What area precision was reflected is degree of closeness of this method identification area relative to quasi-value.Verification method is：With
The cultivated area in the research area in statistical yearbook evaluates this method recognition result as quasi-value.It is small that area precision is defined as the winter
The actual cultivated area of wheat deducts identification area and the ratio after real area difference with real area, and calculation formula is：
；
In formula,AaFor area precision；AFor the research area winter wheat gross area of this method extraction；A O For the total yearbook of winter wheat
Statistics.
The specific the result of positional precision and area precision is as shown in the table.
It is can be seen that from above-mentioned the result using method provided by the present invention for winter wheat planting area and specific
The identification of distributing position has all reached very high precision, more can reflect accurately and that really real winter wheat plants feelings
Condition.
To sum up, the present invention is based on aspect rhythm and pace of moving things method and fuzzy classification technology, by comprehensively utilizing low resolution and middle height
The remote sensing image of resolution ratio, definite spatial distribution result of the winter wheat under the conditions of middle high-resolution is obtained, has been made up simultaneously
Two methods each the shortcomings that, thus remote sensing monitoring in winter wheat, yield assessment etc. have it is more important and actual
Application value, while also provide new reference for the remote sensing monitoring of other agrotypes.In used in the present invention
The remote sensing image data of low resolution yardstick is easily obtained, and is easy to carry out actual monitoring application on certain regional scale, because
And also there is preferable application value.
Claims (4)
1. the winter wheat remote sensing recognition method of the crucial Aspection character of synthesis and fuzzy classification technology, it is characterised in that this method bag
Include following steps：
（1）Data prediction
Data of the TM middle high-resolutions image with MODIS coarse resolutions image in projection and coordinate system are completely registering；
（2）Abundance figure under preparation research area coarse resolution
A. according to research area winter wheat plantation situation, in the case where ensuring precision conditions, typical sample investigation MODIS rough segmentations are laid in field
Winter wheat planting area ratio in one pixel of resolution yardstick, sample is provided for regression analysis；Or utilize high-definition remote sensing
Image is interpreted to winter wheat plantation situation, makes the pixel percent data of MODIS coarse resolution yardsticks based on this
As sample, so as to replace field operation to lay sample prescription；
B. extraction reflects the period in winter wheat typical case's phenological period from MODIS NDVI time serieses, utilizes the MODIS of the period
NDVI data establish typical phenological period slope image；
C. regression model is established with sample data obtained by step A using typical phenological period slope image obtained by step B, and is expanded
Whole research area is opened up, so as to obtain studying the abundance figure of the winter wheat planting area in area；
（3）Degree of membership of the pixel for winter wheat under the middle high-resolution yardstick in acquisition research area
A. according to TM data, using fuzzy classification technology, the degree of membership based on bayesian criterion determines method, obtains each pixel and is subordinate to
Belong to the degree of membership of winter wheat；
B. Data correction, " the different spectrum of jljl " phenomenon is eliminated；
（4）Comprehensive Evaluation, i.e. combining step（2）Abundance figure and step（3）It is subordinate to degrees of data and feelings is planted to winter wheat in remote sensing figure
Condition is judged
A. in each MODIS pixels, the TM pixels corresponding to it are sorted from high to low to the degree of membership of winter wheat；
If B. the winter wheat abundance of a MODIS pixel is F%, then will under the conditions of middle high-resolution, on locus institute it is right
Answer 10 × 10, the preceding F% pixel to be sorted by abovementioned steps A winter wheat degree of membership be determined as winter wheat, F values are according to step
Suddenly（2）In abundance figure determine；
C. whole research area is expanded in this way, you can to obtain the cultivated area of winter wheat and distribution.
2. integrating the winter wheat remote sensing recognition method of crucial Aspection character and fuzzy classification technology as claimed in claim 1, it is special
Sign is, step（1）Described in TM resolution ratio be 25 meters, MODIS resolution ratio be 250 meters；Registration process is specifically, by TM data
Resampling is set to 25 meters, completely registering with 250 meters of data of MODIS in projection and coordinate system so that each MODIS pixel
10 × 10 TM pixels can be corresponded on locus.
3. integrating the winter wheat remote sensing recognition method of crucial Aspection character and fuzzy classification technology as claimed in claim 1, it is special
Sign is, step（2）Described in winter wheat typical case's phenological period be winter wheat period of seedling establishment to winter wheat boot stage.
4. integrating the winter wheat remote sensing recognition method of crucial Aspection character and fuzzy classification technology as claimed in claim 1, it is special
Sign is that fuzzy classification computational methods are as follows：
Assuming that vector in classification image, research is divided intomClass, i.e.,k i , i=1,2 ..., m, according to
Bayesian formula,XBelonged under conditions of appearancek i The ownership probability of class is：,
VectorXIn classificationk i Conditional probability density function be：
,
In formula,P(k i ) it is classificationk i Prior probability；P(X/k i ) it is classificationk i It is middle vector occurXProbability；nIt is characterized space
Dimension；mTo study area's classification number；μ i For classificationk i Training sample mean vector；For classificationk i The association of training sample
Variance matrixDeterminant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510037425.4A CN104615977B (en) | 2015-01-26 | 2015-01-26 | The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510037425.4A CN104615977B (en) | 2015-01-26 | 2015-01-26 | The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104615977A CN104615977A (en) | 2015-05-13 |
| CN104615977B true CN104615977B (en) | 2018-02-06 |
Family
ID=53150414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510037425.4A Active CN104615977B (en) | 2015-01-26 | 2015-01-26 | The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104615977B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915660A (en) * | 2015-07-01 | 2015-09-16 | 中国科学院遥感与数字地球研究所 | Winter wheat automatic recognition method based on GF-1/WFV NDVI time sequence |
| CN104951772B (en) * | 2015-07-10 | 2017-12-29 | 中国科学院遥感与数字地球研究所 | A kind of winter wheat extracting method based on NDVI time-serial positions integration |
| CN107330801A (en) * | 2017-06-07 | 2017-11-07 | 北京师范大学 | The computational methods and device of a kind of winter wheat planting proportion |
| CN109493247A (en) * | 2017-09-11 | 2019-03-19 | 苏州市农业科学院 | The method that confirmation field crop rotation is lain fallow |
| CN108805079B (en) * | 2018-06-12 | 2020-10-09 | 中国科学院地理科学与资源研究所 | Winter wheat identification method and device |
| CN109242875B (en) * | 2018-10-19 | 2020-08-18 | 北京工业职业技术学院 | Winter wheat planting area extraction method and system |
| CN108984803B (en) * | 2018-10-22 | 2020-09-22 | 北京师范大学 | Method and system for spatializing crop yield |
| CN109492568B (en) * | 2018-10-31 | 2021-10-08 | 武汉珈和科技有限公司 | Method and system for extracting planting distribution of global specific crop main producing area |
| CN111709379B (en) * | 2020-06-18 | 2023-04-18 | 广西壮族自治区农业科学院 | Remote sensing image-based hilly area citrus planting land plot monitoring method and system |
| CN112347992B (en) * | 2020-12-01 | 2023-07-18 | 中国林业科学研究院 | Remote sensing estimation method for time sequence AGB in desert area |
| CN114283335B (en) * | 2021-12-27 | 2022-11-22 | 河南大学 | Historical period remote sensing identification precision verification preparation method |
| CN114529097B (en) * | 2022-02-26 | 2023-01-24 | 黑龙江八一农垦大学 | Multi-scale crop phenological period remote sensing dimensionality reduction prediction method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004001659A1 (en) * | 2001-12-17 | 2003-12-31 | Digitalglobe Inc. | System, method, and apparatus for satellite remote sensing |
| CN102346808A (en) * | 2011-06-08 | 2012-02-08 | 北京师范大学 | Method for inverting LAI (leaf area index) from HJ-1 satellite data |
| CN102609726A (en) * | 2012-02-24 | 2012-07-25 | 中国科学院东北地理与农业生态研究所 | Method for classifying remote sensing images blended with high-space high-temporal-resolution data by object oriented technology |
| CN103033150A (en) * | 2011-10-09 | 2013-04-10 | 黄青 | Method for quickly extracting main crop planting area through utilization of moderate resolution imaging spectroradiometer (MODIS) data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336956B (en) * | 2013-07-10 | 2016-08-10 | 福州大学 | A kind of winter wheat area evaluation method based on remote sensing time series data |
-
2015
- 2015-01-26 CN CN201510037425.4A patent/CN104615977B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004001659A1 (en) * | 2001-12-17 | 2003-12-31 | Digitalglobe Inc. | System, method, and apparatus for satellite remote sensing |
| CN102346808A (en) * | 2011-06-08 | 2012-02-08 | 北京师范大学 | Method for inverting LAI (leaf area index) from HJ-1 satellite data |
| CN103033150A (en) * | 2011-10-09 | 2013-04-10 | 黄青 | Method for quickly extracting main crop planting area through utilization of moderate resolution imaging spectroradiometer (MODIS) data |
| CN102609726A (en) * | 2012-02-24 | 2012-07-25 | 中国科学院东北地理与农业生态研究所 | Method for classifying remote sensing images blended with high-space high-temporal-resolution data by object oriented technology |
Non-Patent Citations (2)
| Title |
|---|
| "地块支持下MODIS-NDVI时间序列冬小麦种植面积测量研究";李乐 等;《光谱学与光谱分析》;20110531;第31卷(第5期);第1379-1383页 * |
| "模糊分类技术在作物类型识别中的应用";尤淑撑 等;《国土资源遥感》;20000315(第1期，总第43期);第39-43页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104615977A (en) | 2015-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104615977B (en) | The winter wheat remote sensing recognition method of comprehensive crucial Aspection character and fuzzy classification technology | |
| CN111709379B (en) | Remote sensing image-based hilly area citrus planting land plot monitoring method and system | |
| CN110378909B (en) | Single wood segmentation method for laser point cloud based on Faster R-CNN | |
| CN105608293B (en) | Merge Biomass retrieval method and system on the woodland of spectrum and textural characteristics | |
| Jung et al. | Exploiting synergies of global land cover products for carbon cycle modeling | |
| CN105740759B (en) | Semilate rice information decision tree classification approach based on feature extraction in multi-temporal data | |
| Li et al. | Classification of tree species based on structural features derived from high density LiDAR data | |
| Halme et al. | Utility of hyperspectral compared to multispectral remote sensing data in estimating forest biomass and structure variables in Finnish boreal forest | |
| CN110363246B (en) | Fusion method of vegetation index NDVI with high space-time resolution | |
| CN103063202A (en) | Cyanobacteria biomass spatial-temporal change monitoring and visualization method based on remote sensing image | |
| CN102354328B (en) | Leaf area index (LAI) product inversion method and system for global earth surface | |
| Davidson et al. | A comparison of three approaches for predicting C4 species cover of northern mixed grass prairie | |
| Fei et al. | Remote sensing of mangrove wetlands identification | |
| CN107423537A (en) | A kind of method of the surface temperature NO emissions reduction based on adaptive threshold | |
| CN104794424A (en) | Novel middle-low resolution remote sensing data combined cultivated land identification method | |
| Zhang et al. | Winter wheat identification by integrating spectral and temporal information derived from multi-resolution remote sensing data | |
| CN109460789A (en) | A kind of soil moisture fusion method based on Bayes's maximum entropy | |
| CN109034189A (en) | Forest type identification method based on high score remote sensing image | |
| CN109960972A (en) | A kind of farm-forestry crop recognition methods based on middle high-resolution timing remotely-sensed data | |
| CN116863341B (en) | Crop classification and identification method and system based on time sequence satellite remote sensing image | |
| Vehmas et al. | Identification of boreal forest stands with high herbaceous plant diversity using airborne laser scanning | |
| Silva Junior et al. | Object-based image analysis supported by data mining to discriminate large areas of soybean | |
| Greenberg et al. | A bottom-up approach to vegetation mapping of the Lake Tahoe Basin using hyperspatial image analysis | |
| Yi et al. | Object-oriented remote sensing imagery classification accuracy assessment based on confusion matrix | |
| Li et al. | Performance of GEDI data combined with Sentinel-2 images for automatic labelling of wall-to-wall corn mapping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |