CN109242875A - A kind of winter wheat planting area extracting method and system - Google Patents
A kind of winter wheat planting area extracting method and system Download PDFInfo
- Publication number
- CN109242875A CN109242875A CN201811219180.7A CN201811219180A CN109242875A CN 109242875 A CN109242875 A CN 109242875A CN 201811219180 A CN201811219180 A CN 201811219180A CN 109242875 A CN109242875 A CN 109242875A
- Authority
- CN
- China
- Prior art keywords
- winter wheat
- modis
- area
- data
- winter
- 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
- 241000209140 Triticum Species 0.000 title claims abstract description 461
- 235000021307 Triticum Nutrition 0.000 title claims abstract description 461
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000005259 measurement Methods 0.000 claims abstract description 40
- 238000012952 Resampling Methods 0.000 claims abstract description 24
- 230000003595 spectral effect Effects 0.000 claims abstract description 24
- 238000001228 spectrum Methods 0.000 claims description 30
- 238000000605 extraction Methods 0.000 claims description 25
- 238000012937 correction Methods 0.000 claims description 24
- 239000000284 extract Substances 0.000 claims description 18
- 238000000354 decomposition reaction Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000002310 reflectometry Methods 0.000 claims description 10
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 claims description 9
- 238000000926 separation method Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000013481 data capture Methods 0.000 claims description 5
- 235000013399 edible fruits Nutrition 0.000 claims description 5
- 238000002329 infrared spectrum Methods 0.000 claims description 5
- 238000004566 IR spectroscopy Methods 0.000 claims description 4
- 238000002156 mixing Methods 0.000 claims description 4
- 238000013316 zoning Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000003973 irrigation Methods 0.000 description 4
- 230000002262 irrigation Effects 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000003909 pattern recognition Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 2
- 230000000052 comparative effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009394 selective breeding Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 244000025254 Cannabis sativa Species 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009313 farming Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003472 neutralizing effect Effects 0.000 description 1
- 238000010899 nucleation Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010200 validation analysis Methods 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
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Probability & Statistics with Applications (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of winter wheat planting area extracting method and system, this method includes obtaining TM data and MODIS data;TM data are classified using K-means non-supervised classification to obtain TM recognition result;MODIS data are decomposed using linear spectral unmixing model to obtain MODIS recognition result;Resampling TM data;Determine TM typical case winter wheat area and TM atypia winter wheat area;Determine consistency and nonconforming demarcation threshold;There are non-wheat mistakes to enter, and corrects TM recognition result using MODIS recognition result;There are winter wheat mistakes to go out, and corrects TM recognition result using MODIS recognition result;The cultivated area of final winter wheat is obtained in conjunction with the crop coverage measurement result of the first winter wheat and the crop coverage measurement result of the second winter wheat.The TM winter wheat image element information in the winter wheat Information revision nonuniformity region that the present invention is extracted using MODIS time series, improves the measurement accuracy of winter wheat planting area.
Description
Technical field
The present invention relates to crops studying technological domains, more particularly to a kind of winter wheat planting area extracting method and are
System.
Background technique
The information extraction of crop planting range and spatial distribution is one of remote sensing technology important applied field.Farming species
It is numerous to plant area extraction method, according to the difference of image resolution, is divided into measurement method that is high, neutralizing three scales of low point of rate.
High-resolution remote sensing image (including QUICKBIRD, IKONOS etc.) mainly takes the classification side of digitizing solution and object-oriented
Method;Intermediate-resolution remote sensing image (including TM, IRS-P6, SPOT etc.) mainly takes non-supervisory, supervised classification method and mixing picture
First decomposition method;Low resolution remote sensing image (including MODIS, NOAA) mostly uses Decomposition of Mixed Pixels and pattern-recognition side
Method.Three scale crop information extracting method features are as follows: 1) high-resolution remote sensing image, due to the increase of spatial information,
It is disturbed increase, brings very big difficulty to classification, manual digitalization is then time-consuming, laborious.Furthermore high-definition picture price
Valuableness can not carry out large-scale crop acreage monitoring, therefore high-resolution is only used for the information of small range crops
It extracts.2) intermediate-resolution remote sensing image coverage area is wider, and the return visit period is shorter, relatively crop acreage remote sensing is suitble to survey
Amount.But " foreign matter is with spectrum, the different spectrum of jljl " phenomenon is necessarily caused using single phase remote sensing image, to the information extraction band of crops
Carry out very big difficulty;And it is although able to solve this problem using more phase remote sensing images, but the medium resolution satellite return visit period is longer,
In addition the influence of the weather conditions such as cloud, it is difficult to more phase images are obtained within crop growth season.3) low resolution remote sensing image has
The characteristics of having time sequence, can effectively depict the growth cycle of crops, therefore can be by effectively distinguishing Different Crop.
But since the pixel resolution of low resolution remote sensing images is lower, mixed pixel phenomenon is serious, and nicety of grading is not high.
To sum up, how to provide a kind of technical solution that can extract crop acreage on a large scale and accurately is this field skill
The problem of art personnel's urgent need to resolve.
Summary of the invention
The object of the present invention is to provide a kind of winter wheat planting area extracting method and systems, to solve intermediate-resolution remote sensing
" foreign matter is with spectrum, the different spectrum of jljl " phenomenon present in Extraction of Image method leads to the problem of crop coverage measurement inaccuracy.
To achieve the above object, the present invention provides a kind of winter wheat planting area extracting methods, this method comprises:
Obtain TM data and MODIS data;The TM data are what Medium resolution remotely sensed data was obtained through data prediction
Data, the MODIS data are the data that low resolution remote sensing images are obtained through data prediction;
The TM data are classified using K-means non-supervised classification to obtain TM recognition result, the TM knows
Other result includes TM winter wheat area, the non-winter wheat area TM and TM winter wheat pixel abundance figure;
The MODIS data are decomposed using linear spectral unmixing model to obtain MODIS recognition result, it is described
MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat time series data and MODIS
Winter wheat pixel abundance figure;
Resampling is carried out in default sample range, obtains resampling TM data;
TM typical case winter wheat area and TM atypia winter wheat area are determined according to the resampling TM data;
According to TM winter wheat area and the non-winter wheat area TM and MODIS winter wheat area and the non-winter wheat of MODIS
Area carries out consistency analysis, determines consistency and nonconforming demarcation threshold;
It in TM typical case winter wheat area, judges whether there is non-wheat mistake and enters, if so, described in being utilized according to time series
MODIS recognition result corrects the TM recognition result, obtains the crop coverage measurement result of the first winter wheat;
It in the TM atypia winter wheat area, judges whether there is winter wheat mistake and goes out, if so, utilizing institute according to time series
It states MODIS recognition result and corrects the TM recognition result, obtain the crop coverage measurement result of the second winter wheat;
In conjunction with the crop coverage measurement result of first winter wheat and the crop coverage measurement knot of second winter wheat
Fruit obtains the cultivated area of final winter wheat.
Optionally, the acquisition TM data and MODIS data, specifically include
Geometric correction is carried out to the Medium resolution remotely sensed data using quadratic polynomial and bilinear interpolation method, obtains TM
Data;
Cloud removing is carried out to the low resolution remote sensing images, the MODIS image after obtaining cloud;
Orthogonal vegetation index is calculated using formula PVI=NIR-RED, obtains MODIS data, the MODIS data include
Orthogonal vegetation index with time series;Wherein, PVI is orthogonal vegetation index, and NIR is the reflectivity of near infrared spectrum, RED
For the reflectivity of infrared spectroscopy.
Optionally in described to be decomposed to obtain MODIS to the MODIS data using linear spectral unmixing model
Recognition result specifically includes:
The extraction of winter wheat image element information is carried out using linear spectral unmixing model, obtains MODIS winter wheat pixel letter
Breath;
According to the orthogonal vegetation index and time series of the MODIS data, it is bent to obtain typical feature time series feature
Line;
According to the typical feature time series indicatrix, endmember is selected by auxiliary data of the TM data,
Classified using minimal noise separation conversion method to the MODIS winter wheat image element information, obtains MODIS recognition result;
The MODIS recognition result include MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat time series data and
MODIS winter wheat pixel abundance figure, wherein MODIS winter wheat area and the non-winter wheat area the MODIS areWherein, when in typical feature time series indicatrix
Third wave band orthogonal vegetation index b3Vegetation index b orthogonal with second band2Difference be greater than the 0 and the 18th wave band
Orthogonal vegetation index b18Vegetation index b orthogonal with the 16th wave band16Difference less than 0 when, Pm=1 is MODIS winter wheat area;
Other situations, Pm=0 is the non-winter wheat area MODIS.
Optionally, it is described according to TM winter wheat area and the non-winter wheat area TM and MODIS winter wheat area and
The non-winter wheat area MODIS carries out consistency analysis, determines consistency and nonconforming demarcation threshold, specifically includes:
Extract true winter wheat area in TM winter wheat area;
Obtain region recognition result identical with the true winter wheat area in the MODIS winter wheat area;
The decomposition essence of MODIS data mixing pixel is calculated according to the true winter wheat area and the region recognition result
Degree;
Interval range where determining the abundance difference of winter wheat true value and MODIS recognition result according to the Decomposition Accuracy
Interior pixel accounts for the maximum ratio of whole collecting samples;
Demarcation threshold is determined according to the interval range.
Optionally, described in TM typical case winter wheat area, it judges whether there is non-wheat mistake and enters, if so, according to the time
Sequence corrects the TM recognition result using the MODIS recognition result, obtain the crop coverage measurement of the first winter wheat as a result,
It specifically includes;
It judges whether there is non-winter wheat and is identified as winter wheat, if so, determining that there are non-wheat mistakes to enter;Utilize formulaCalculate the winter wheat pixel in TM typical case winter wheat area
Information, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRFor the winter wheat abundance after correction, T is point
Boundary's threshold value is 40%;
If it is not, then the winter wheat image element information in TM typical case winter wheat area is TM winter wheat abundance.
Optionally, described to judge whether there is winter wheat mistake in the TM atypia winter wheat area and go out, if so, according to when
Between sequence using the MODIS recognition result correct the TM recognition result, obtain the crop coverage measurement knot of the second winter wheat
Fruit specifically includes:
It judges whether there is part winter wheat and is identified as non-winter wheat, if so, determining that there are winter wheat mistakes to go out, utilize
FormulaCalculate the winter wheat picture in the nonuniformity region
Metamessage, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRFor the winter wheat abundance after correction, T is
Demarcation threshold is 40%;
If it is not, then the winter wheat image element information in the TM atypia winter wheat area is TM winter wheat abundance.
The present invention also provides a kind of winter wheat planting area extraction system, which includes:
Data capture unit, for obtaining TM data and MODIS data;The TM data are Medium resolution remotely sensed data warp
The data that data prediction obtains, the MODIS data are the data that low resolution remote sensing images are obtained through data prediction;
TM recognition unit obtains TM knowledge for being classified using K-means non-supervised classification to the TM data
Not as a result, the TM recognition result includes TM winter wheat area, the non-winter wheat area TM and TM winter wheat pixel abundance figure;
MODIS recognition unit, for being decomposed to obtain to the MODIS data using linear spectral unmixing model
MODIS recognition result, the MODIS recognition result include MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat
Time series data and MODIS winter wheat pixel abundance figure;
Resampling unit obtains resampling TM data for carrying out resampling in default sample range;
Zoning unit, for determining TM typical case winter wheat area and TM atypia winter wheat according to the resampling TM data
Area;
Demarcation threshold determination unit, for according to TM winter wheat area and the non-winter wheat area TM and the MODIS winter
The small area of wheat and the non-winter wheat area MODIS carry out consistency analysis, determine consistency and nonconforming demarcation threshold;
Mistake enters amending unit, for judging whether there is non-wheat mistake and entering, if so, root in TM typical case winter wheat area
The TM recognition result is corrected using the MODIS recognition result according to time series, the cultivated area for obtaining the first winter wheat mentions
Take result;
Mistake goes out amending unit, for judging whether there is winter wheat mistake and going out in the TM atypia winter wheat area, if so,
The TM recognition result is corrected using the MODIS recognition result according to time series, obtains the cultivated area of the second winter wheat
Extract result;
Combining unit, the kind for crop coverage measurement result and second winter wheat in conjunction with first winter wheat
It plants area extraction result and obtains the cultivated area of final winter wheat.
Optionally, the data capture unit includes:
Geometric correction subelement, for utilizing quadratic polynomial and bilinear interpolation method to the Medium resolution remotely sensed data
Geometric correction is carried out, TM data are obtained;
Cloud subelement is removed, the MODIS figure for carrying out cloud removing to the low resolution remote sensing images, after obtaining cloud
Picture;
Orthogonal vegetation index computation subunit is obtained for calculating orthogonal vegetation index using formula PVI=NIR-RED
MODIS data, the MODIS data include the orthogonal vegetation index with time series;Wherein, PVI is orthogonal vegetation index,
NIR is the reflectivity of near infrared spectrum, and RED is the reflectivity of infrared spectroscopy.
Optionally, the MODIS recognition unit specifically includes:
Image element information extracts subelement, mentions for carrying out winter wheat image element information using linear spectral unmixing model
It takes, obtains MODIS winter wheat image element information;
Indicatrix determines subelement, for the orthogonal vegetation index and time series according to the MODIS data, obtains
Typical feature time series indicatrix;
Classification subelement, is used for according to the typical feature time series indicatrix, using the TM data as supplementary number
According to selection endmember, is classified using minimal noise separation conversion method to the MODIS winter wheat image element information, obtained
MODIS recognition result;The MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat
Time series data and MODIS winter wheat pixel abundance figure, wherein MODIS winter wheat area and the non-winter wheat of the MODIS
Qu WeiWherein, when typical feature time series feature is bent
The orthogonal vegetation index b of third wave band in line3Vegetation index b orthogonal with second band2Difference be greater than the 0 and the 18th wave
The orthogonal vegetation index b of section18Vegetation index b orthogonal with the 16th wave band16Difference less than 0 when, Pm=1 is small for the MODIS winter
The area of wheat;Other situations, Pm=0 is the non-winter wheat area MODIS.
Optionally, the mistake enters amending unit and is identified as winter wheat for judging whether there is non-winter wheat, if so,
Determine that there are non-wheat mistakes to enter;Utilize formulaDescribed in calculating
The winter wheat image element information in TM typical case winter wheat area, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PR
For the winter wheat abundance after correction, T is demarcation threshold, is 40%;If it is not, the then winter wheat pixel in TM typical case winter wheat area
Information is TM winter wheat abundance;
The wrong amending unit out is identified as non-winter wheat for judging whether there is part winter wheat, if so, really
Surely there is winter wheat mistake to go out, utilize formulaIt calculates described non-
The winter wheat image element information of Uniform Domains, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRFor school
Winter wheat abundance after just, T are demarcation threshold, are 40%;If it is not, then the winter wheat pixel in the TM atypia winter wheat area is believed
Breath is TM winter wheat abundance.
The specific embodiment provided according to the present invention, the invention discloses following technical effects: in the prior art merely with
Intermediate-resolution remote sensing image (TM) carries out winter wheat information extraction, due to being influenced by " foreign matter is with spectrum, the different spectrum of jljl " phenomenon,
Generate winter wheat information mistake enter, it is wrong occur as.Single phase remote sensing image not can avoid the generation of this phenomenon, so as to cause winter wheat
Crop coverage measurement precision reduce.Low resolution has been used in winter wheat planting area extracting method and system provided by the invention
Rate remote sensing image (MODIS), low resolution remote sensing image (MODIS) have time series data, can clearly reflect that the winter is small
Wheat is different from the temporal characteristics curve of other atural objects, is based on MODIS data time series characteristic, utilizes MODIS mixed pixel point
Solve modified result TM data " mistake enters, mistake goes out " phenomenon and caused by error in classification, MODIS winter wheat measurement result auxiliary under,
It realizes and TM data winter wheat measurement result is effectively corrected, improve the measurement accuracy of winter wheat.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of winter wheat planting area extracting method provided in an embodiment of the present invention;
Fig. 2 is the block diagram of winter wheat planting area extraction system provided in an embodiment of the present invention;
Fig. 3 is the spectral curve of typical feature on TM image;
Fig. 4 is several typical feature spectral signature curve graphs of MODIS data;
Fig. 5 is the measurement result figure that MODIS extracts winter wheat, and wherein a is MODIS endmember scatter plot;B is
MODIS winter wheat measurement result;
Fig. 6 is that MODIS-TM winter wheat measures consistency analysis result figure, wherein a is TM/MODIS measurement result difference
Histogram;B is TM/MODIS measurement result difference accumulation histogram;
Fig. 7 is that MODIS amendment TM winter wheat mistake goes out comparative result figure, wherein a is picture before correcting;B is to scheme after correcting
Piece;
Fig. 8 is that MODIS amendment TM winter wheat mistake enters comparative result figure, wherein a is picture before correcting;B is to scheme after correcting
Piece.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of winter wheat planting area extracting method and systems, to solve intermediate-resolution remote sensing
" foreign matter is with spectrum, the different spectrum of jljl " phenomenon present in Extraction of Image method leads to the problem of crop coverage measurement inaccuracy.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
As shown in Figure 1, winter wheat planting area extracting method provided by the invention, comprising:
Step 101: obtaining TM data and MODIS data;The TM data are that Medium resolution remotely sensed data is located in advance through data
Obtained data are managed, the MODIS data are the data that low resolution remote sensing images are obtained through data prediction.
By intermediate-resolution remote sensing image equipment and low resolution remote sensing image equipment shoot Medium resolution remotely sensed data and
Low resolution remote sensing images, this partial graphical are several spectrum pictures in continuous time period, the more difference of time and season
The spectrum pel data presented in image is also just different.
It is specifically included to step 101:
Geometric correction is carried out to the Medium resolution remotely sensed data using quadratic polynomial and bilinear interpolation method, obtains TM
Data;
Cloud removing is carried out to the low resolution remote sensing images, the MODIS image after obtaining cloud;
Orthogonal vegetation index is calculated using formula PVI=NIR-RED, obtains MODIS data, the MODIS data include
Orthogonal vegetation index with time series;Wherein, PVI is orthogonal vegetation index, and NIR is the reflectivity of near infrared spectrum, RED
For the reflectivity of infrared spectroscopy.
Step 102: the TM data are classified to obtain TM recognition result using K-means non-supervised classification,
The TM recognition result includes TM winter wheat area, the non-winter wheat area TM and TM winter wheat pixel abundance figure.
Step 103: the MODIS data being decomposed using linear spectral unmixing model to obtain MODIS identification
As a result, the MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat time series number
According to MODIS winter wheat pixel abundance figure.
The step 103 specifically includes:
The extraction of winter wheat image element information is carried out using linear spectral unmixing model, obtains MODIS winter wheat pixel letter
Breath;
According to the orthogonal vegetation index and time series of the MODIS data, it is bent to obtain typical feature time series feature
Line;
According to the typical feature time series indicatrix, endmember is selected by auxiliary data of the TM data,
Classified using minimal noise separation conversion method to the MODIS winter wheat image element information, obtains MODIS recognition result;
The MODIS recognition result include MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat time series data and
MODIS winter wheat pixel abundance figure, wherein MODIS winter wheat area and the non-winter wheat area the MODIS areWherein, when in typical feature time series indicatrix
Third wave band orthogonal vegetation index b3Vegetation index b orthogonal with second band2Difference be greater than the 0 and the 18th wave band
Orthogonal vegetation index b18Vegetation index b orthogonal with the 16th wave band16Difference less than 0 when, Pm=1 is MODIS winter wheat area;
Other situations, Pm=0 is the non-winter wheat area MODIS.
Step 104: carrying out resampling in default sample range, obtain resampling TM data.
Step 105: determining TM typical case winter wheat area and TM atypia winter wheat area according to the resampling TM data;
Step 106: according to TM winter wheat area and the non-winter wheat area TM and MODIS winter wheat area and MODIS
Non- winter wheat area carries out consistency analysis, determines consistency and nonconforming demarcation threshold.
The step 106 specifically includes:
Extract true winter wheat area in TM winter wheat area;
Obtain region recognition result identical with the true winter wheat area in the MODIS winter wheat area;
The decomposition essence of MODIS data mixing pixel is calculated according to the true winter wheat area and the region recognition result
Degree;
Interval range where determining the abundance difference of winter wheat true value and MODIS recognition result according to the Decomposition Accuracy
Interior pixel accounts for the maximum ratio of whole collecting samples;
Demarcation threshold is determined according to the interval range.
Step 107: in TM typical case winter wheat area, judging whether there is non-wheat mistake and enter, if so, according to time series
The TM recognition result is corrected using the MODIS recognition result, obtains the crop coverage measurement result of the first winter wheat;
Specifically, judging whether there is non-winter wheat is identified as winter wheat, if so, determining that there are non-wheat mistakes to enter;
Utilize formulaCalculate the winter in TM typical case winter wheat area
Wheat image element information, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRIt is rich for the winter wheat after correction
Degree, T is demarcation threshold, is 40%;
If it is not, then the winter wheat image element information in TM typical case winter wheat area is TM winter wheat abundance.
Step 108: in the TM atypia winter wheat area, judging whether there is winter wheat mistake and go out, if so, according to time sequence
Column correct the TM recognition result using the MODIS recognition result, obtain the crop coverage measurement result of the second winter wheat;
Specifically, judging whether there is part winter wheat is identified as non-winter wheat, if so, determining that there are winter wheat mistakes
Out, formula is utilizedCalculate the winter in the nonuniformity region
Wheat image element information, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRIt is rich for the winter wheat after correction
Degree, T is demarcation threshold, is 40%;
If it is not, then the winter wheat image element information in the TM atypia winter wheat area is TM winter wheat abundance.
Step 109: in conjunction with the crop coverage measurement result of first winter wheat and the growing surface of second winter wheat
Product extracts result and obtains the cultivated area of final winter wheat.
As shown in Fig. 2, winter wheat growing surface corresponding with above-mentioned winter wheat planting area extracting method provided by the invention
Product extraction system, comprising:
Data capture unit 201, for obtaining TM data and MODIS data;The TM data are intermediate-resolution remote sensing figure
As the data obtained through data prediction, the MODIS data are the number that low resolution remote sensing images are obtained through data prediction
According to;
TM recognition unit 202, for being classified to obtain TM to the TM data using K-means non-supervised classification
Recognition result, the TM recognition result include TM winter wheat area, the non-winter wheat area TM and TM winter wheat pixel abundance figure;
MODIS recognition unit 203, for being decomposed using linear spectral unmixing model to the MODIS data
MODIS recognition result is obtained, the MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter
Wheat time series data and MODIS winter wheat pixel abundance figure;
Resampling unit 204 obtains resampling TM data for carrying out resampling in default sample range;
Zoning unit 205, for determining that TM typical case winter wheat area and TM atypia winter are small according to the resampling TM data
The area of wheat;
Demarcation threshold determination unit 206, for according to TM winter wheat area and the non-winter wheat area TM and the MODIS
Winter wheat area and the non-winter wheat area MODIS carry out consistency analysis, determine consistency and nonconforming demarcation threshold;
Mistake enters amending unit 207, for judging whether there is non-wheat mistake and entering in TM typical case winter wheat area, if so,
The TM recognition result is corrected using the MODIS recognition result according to time series, obtains the cultivated area of the first winter wheat
Extract result;
Mistake goes out amending unit 208, for judging whether there is winter wheat mistake and going out in the TM atypia winter wheat area, if
It is that the TM recognition result is corrected using the MODIS recognition result according to time series, obtains the growing surface of the second winter wheat
Product extracts result;
Combining unit 209, in conjunction with first winter wheat crop coverage measurement result and second winter wheat
Crop coverage measurement result obtain the cultivated area of final winter wheat.
It should be noted that above each unit further includes respective subelement, record in front, it is no longer superfluous herein
It states.
Above content is explained in detail below with reference to a specific embodiment.
One, embodiment research area
The research area of selection be located at Beijing, Tianjin, Hebei province have a common boundary, range be 116 ° of 1 ' 11 " 33 ' 4 " E of E -117 °, 39 °
41'12"N—40°29'27".The region based on Plain, the north be Yanshan mountain range, east large area Winter Wheat Planted, therebetween
It is miscellaneous to have large stretch of bare area, water body.
Two, embodiment data
1, TM data
Select the Landsat TM image on April 7th, 2006, orbit number 123/32.TM data prediction mainly includes
Geometry accurate correct.On the basis of standard video figure in 1999, using quadratic polynomial and bilinear interpolation method to research area TM
Image carries out the geometric correction of image to image, examines through reconnaissance again, error is in a pixel.Due to being to utilize single phase shadow
As classification progress information extraction, there is no need to carry out atmospheric correction.
The period is in the winter wheat jointing stage, with the spectral information of artificial pasture close to (such as golf course, urban afforestation
Meadow etc.).In the winter wheat jointing stage, wheat irrigation is relatively more, and the winter wheat plot after irrigation is mixed with the spectral information of water body, this
It will lead to winter wheat and be easily accidentally divided into non-winter wheat.The curve of spectrum of all kinds of atural objects can be seen that winter wheat and meadow from Fig. 3
Relatively, the winter wheat of irrigation is influenced curve by background water spectral, and spectral information tends to water body.Through above-mentioned point
Analysis, single phase Remote Spectra information are difficult to the mixed atural object of effective district split-phase.
2, MODIS data
There are many types, including red, nearly red, NDVI, EVI etc. for MODIS data.The present embodiment was synthesized using MODIS 16 days
Data (about 2 issue evidence in January, totally 18 phase), resolution ratio 250m, time are first arrival in October, 2005 in late June, 2006, across
The growth cycle of more entire winter wheat.Orthogonal vegetation index (perpendicularvegetation index, PVI) can have
Effect carries out linear decomposition, sees formula (1).
PVI=NIR-RED (1)
Wherein, NIR represents near-infrared, and RED represents feux rouges.
Three, winter wheat planting area extraction process
1, data prediction
Medium resolution remotely sensed data pretreatment mainly includes that stringent geometric correction and research area extract.With standard in 1999
On the basis of striograph, image is carried out to figure to research area's Medium resolution remotely sensed data using quadratic polynomial and bilinear interpolation method
The geometric correction of picture, is examined through reconnaissance again, and error obtains TM data in a pixel.
Low resolution preprocessing of remote sensing images mainly includes that cloud removing and PVI are calculated.Low point on inspection in the present embodiment
Resolution remote sensing images are intact in each period quality, and there is no the influences of cloud, therefore directly calculate PVI by formula (1).
2, TM data extract winter wheat
Classification system is defined as 4 kinds of atural objects, respectively winter wheat, water body, cities and towns/bare area, massif.Wherein, due to city
Town is easy mutually to mix with bare area spectral information, therefore is a seed type by two kinds of terrestrial object information merger.From the point of view of classification results, big portion
Divide winter wheat to be extracted efficiently out, but still have the following problems: 1) it is accurate to carry out the similar meadow of spectrum and winter wheat
It distinguishes;2) winter wheat and the bad winter wheat of growing way after irrigating can not be identified by standard.By the water body in recognition result, cities and towns/naked
Ground and massif merger are non-winter wheat area (being denoted as Pt).Based on 896 field GPS measurement point datas, TM image is identified
As a result precision evaluation (being shown in Table 1) is carried out.Precision evaluation the results show that winter wheat nicety of grading be 88.39%, show TM data energy
Enough effective information extractions for carrying out winter wheat.But the mistake of winter wheat information extraction enters, wrong region out does not have too many field data
Support, influence that " foreign matter with spectrum, the different spectrum of jljl " phenomenon extracts result to winter wheat can not be measured, these phenomenons are to a certain degree
On by visual observation interpretation to be judged.
1 TM winter wheat nicety of grading of table
3, MODIS data winter wheat is extracted
Studying area mainly includes three kinds of vegetation: winter wheat, meadow (artificial pasture/natural meadow) and the woods, wherein with the winter
The green vegetation that wheat mutually mixes is mainly artificial pasture.Study several typical feature curve of spectrum features of area MODIS data such as
Shown in Fig. 4, the wave band corresponding period is as follows: October: 1-2 wave band;November: 3-4 wave band;December: 5-6 wave band;January: 7-8 wave
Section;2 months: 9-10 wave band;March: 11-12 wave band;April: 13-14 wave band;May: 15-16 wave band;June: 17-18 wave band.
Find out from MODIS time-serial position feature, all kinds of atural objects have time sequence that is apparent, being different from other atural objects
Column feature.In the research area, winter wheat is relatively high in PVI value at the beginning of 10 months, declines rapidly at the end of month, into November (seeding stage)
PVI starts to increase, and the coming year 2 months November-, (Wintering Period) PVI value was lower, and second year March (period of seedling establishment) PVI value increases, and is continued for
To the 5-6 month, then reduce rapidly (after harvesting)
Using MODIS data, winter wheat extraction is carried out using linear spectral unmixing model.To improve accuracy,
On the basis of analyzing typical feature time series feature, endmember is selected using TM data as auxiliary data, carries out MNF (most
Small noise separation) transformation (first band is combined with second band, as shown in Fig. 5 (a)).Shown in recognition result such as Fig. 5 (b), greatly
Piece winter wheat is extracted efficiently out;And identification error is mainly distributed on mountain area and (carries out visual interpretation by TM data, discovery should
Region does not have winter wheat distribution) and cities and towns inside, mainly generated by green vegetation inside massif and cities and towns.
4, TM and MODIS winter wheat recognition result consistency analysis
The precision for the winter wheat recognition result that MODIS data extract, be determine its can to TM data " foreign matter with spectrum,
The key that the different spectrum of jljl " problem is modified.Certain area (3269 pixels, winter wheat measurement result are selected from TM data
Precise area) visual interpretation is as truthful data, and with MODIS recognition result progress consistency analysis, calculation method is as follows:
Wherein,It is TM winter wheat true value, aiIt is the winter wheat recognition result of MODIS data linear decomposition, n is sample area
Pixel number.RMSE can reflect the Decomposition Accuracy of mixed pixel on the whole.
From fig. 6, it can be seen that the RMSE of MODIS recognition result is 0.207 in validation region.Wherein, true value and MODIS
Abundance difference ratio shared by the pixel of 0%-30% interval range of recognition result is maximum, accounts for one-hundred-percent inspection sample
88%, illustrate that MODIS and TM recognition result have very strong consistency, which is defined as Uniform Domains;Abundance is poor
Region of the value greater than 40% is defined as inconsistent region.Therefore, it is set as MODIS and TM for 40% and identifies winter wheat result one
It causes, the demarcation threshold T in non-uniform region.
5, TM winter wheat measurement result is corrected
It can be seen that artificial pasture and the spectrum of winter wheat letter in research area in conjunction with TM recognition result and visual interpretation result
Breath is very close, and large stretch of artificial pasture is divided into winter wheat;The spectral information in cities and towns is presented in the winter wheat just irrigated, and causes
Large stretch of winter wheat is divided into city.The problem of these " foreign matter is with spectrum, the different spectrum of jljl " phenomenons lead to misclassification can be small by the winter
The MODIS time series data of wheat is modified.The specific method is as follows:
(1) mistake of winter wheat result goes out to correct
Winter wheat is identified by TM data as a result, its mistake goes out the winter wheat spectral signature and typical case's winter wheat spectrum in region
Feature is different, thus is divided into other atural objects (such as the winter wheat area irrigated).It can be corrected by MODIS data recognition result
TM data identify that winter wheat mistake goes out as a result, its pattern definition is as follows:
Wherein, PMODISFor the winter wheat abundance that MODIS is extracted, PTMFor TM data resampling to 250 meters of winter wheat abundance, PR
For the winter wheat abundance after pattern-recognition, T represents demarcation threshold, is 40%.
According to formula (7), go out region with the winter wheat mistake of MODIS data recognition result amendment TM identification, as a result such as Fig. 7 institute
Show.R is a sub-district for studying area, which has large stretch of irrigation in winter wheat situation (TM visual interpretation), therefore has concentrated TM
The mistake of data identification goes out result.
The precision that TM data mistake goes out result, 50 pixels of artificial selection in R are corrected for verifying MODIS data recognition result
For accuracy detection.Due to atural object pours, it is 0%-20% that TM, which extracts winter wheat abundance, and average abundance 8.6% is deposited
Apparent winter wheat mistake occur as.After MODIS winter wheat recognition result modified R area TM data recognition result, 50 inspections
The abundance range for testing winter wheat measurement result a little is 70.2-100%, average abundance 84.9%.Come from precision evaluation result
It sees and (is shown in Table 2), RMSE is reduced to 0.142 from 0.758, and identification error significantly reduces, and correction effect is obvious.
(2) winter wheat result mistake enters amendment
It studies in area, artificial pasture is manually nursed, and October, PVI value was begun to decline, and starts to return in the March in the coming year
Blueness, it is in rising trend always, it is different from the curve of spectrum trend of winter wheat, it can be reflected by MODIS time series data
Come, but for phase image mono- for TM, the two spectral information is extremely similar, and artificial pasture is easy to be divided into winter wheat.
The pattern definition for entering result with MODIS data recognition result substitution TM data identification winter wheat mistake is as follows:
Wherein, PMODISFor the winter wheat abundance that MODIS is extracted, PTMFor TM data resampling to 250m winter wheat abundance, Ps
For the abundance of the winter wheat after pattern-recognition.As shown in figure 8, in TM typical case winter wheat, the MODIS data correction TM data winter
The result of wheat.
Within the scope of S, due to the limitation of the mono- phase image of TM, artificial pasture is largely divided into winter wheat, exists big
Amount TM data recognition result mistake enters phenomenon.The precision that TM data mistake goes out result is corrected for verifying MODIS data recognition result, in S
50 pixels of interior artificial selection are used for accuracy detection.Precision test result (being shown in Table 2) display, passes through MODIS data recognition result
After the mistake of amendment TM data identification enters result, the abundance range of the winter wheat recognition result of 50 check points is 0%-35% in S,
Average abundance, which is 16.3%, RMSE, drops to 0.122 from 0.901, and identification error significantly reduces, and correction effect is obvious.In Artificial grass
Within the scope of ground, winter wheat true value percentage is 0%.
2 TM winter wheat mistake in/out region MODIS winter wheat measurement result RMSE of table
Four, conclusion
(1) compared with the preferable TM winter wheat recognition result of the quality of data, RMSE is MODIS data winter wheat recognition result
0.207, it is seen that TM and MODIS data winter wheat recognition result comparison of coherence is high, illustrates that MODIS data can effectively be extracted the winter
The cultivated area of wheat.
(2) TM data will lead to the reduction of winter wheat information extraction precision due to " foreign matter with spectrum, the different spectrum of jljl " phenomenon.
The mistake that MODIS data winter wheat recognition result can be used to correct TM data winter wheat enters, mistake goes out recognition result.From correction result
From the point of view of, " the different spectrum of jljl " region (i.e. winter wheat identifies " mistake goes out " region), TM data are corrected by MODIS data recognition result
After recognition result, RMSE drops to 0.142 from 0.758;In winter wheat " foreign matter is with spectrum " region (i.e. winter wheat " mistake enters " region), lead to
After crossing MODIS data recognition result amendment TM data recognition result, RMSE drops to 0.122 from 0.901.As it can be seen that although MODIS is repaired
There are still certain errors for result after just, but relative to TM data recognition result from the point of view of, can greatly improve to a certain extent
The measurement accuracy of winter wheat.
It is led the experimental results showed that MODIS data are able to solve TM data to a certain extent since single phase image limits to
It causes the mistake of winter wheat information extraction to enter, wrong occur as but there is also some shortcomings for this method: first, small using the MODIS data winter
Wheat recognition result is corrected to carry out TM data winter wheat recognition result, due to data resolution difference, will receive TM and MODIS number
The influence of registration accuracy between.Second, MODIS time series is 16 days generated datas, PVI fluctuates larger, MODIS resolution ratio
Low, linear decomposition error is inevitable, and it is good lower than picture quality that this will lead to the winter wheat planting area precision that MODIS is extracted
The TM image of (" foreign matter is with spectrum, the different spectrum of jljl " phenomenon do not occur).
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part
It is bright.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand method and its core concept of the invention;At the same time, for those skilled in the art, foundation
Thought of the invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (10)
1. a kind of winter wheat planting area extracting method characterized by comprising
Obtain TM data and MODIS data;The TM data are the data that Medium resolution remotely sensed data is obtained through data prediction,
The MODIS data are the data that low resolution remote sensing images are obtained through data prediction;
The TM data are classified using K-means non-supervised classification to obtain TM recognition result, the TM identification knot
Fruit includes TM winter wheat area, the non-winter wheat area TM and TM winter wheat pixel abundance figure;
The MODIS data are decomposed using linear spectral unmixing model to obtain MODIS recognition result, it is described
MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat time series data and MODIS
Winter wheat pixel abundance figure;
Resampling is carried out in default sample range, obtains resampling TM data;
TM typical case winter wheat area and TM atypia winter wheat area are determined according to the resampling TM data;
According to TM winter wheat area and the non-winter wheat area TM and MODIS winter wheat area and the non-winter wheat area MODIS into
Row consistency analysis determines consistency and nonconforming demarcation threshold;
It in TM typical case winter wheat area, judges whether there is non-wheat mistake and enters, if so, described in being utilized according to time series
MODIS recognition result corrects the TM recognition result, obtains the crop coverage measurement result of the first winter wheat;
It in the TM atypia winter wheat area, judges whether there is winter wheat mistake and goes out, if so, described in being utilized according to time series
MODIS recognition result corrects the TM recognition result, obtains the crop coverage measurement result of the second winter wheat;
It is obtained in conjunction with the crop coverage measurement result of first winter wheat and the crop coverage measurement result of second winter wheat
To the cultivated area of final winter wheat.
2. winter wheat planting area extracting method according to claim 1, which is characterized in that the acquisition TM data and
MODIS data, specifically include
Geometric correction is carried out to the Medium resolution remotely sensed data using quadratic polynomial and bilinear interpolation method, obtains TM number
According to;
Cloud removing is carried out to the low resolution remote sensing images, the MODIS image after obtaining cloud;
Orthogonal vegetation index is calculated using formula PVI=NIR-RED, obtains MODIS data, the MODIS data include having
The orthogonal vegetation index of time series;Wherein, PVI is orthogonal vegetation index, and NIR is the reflectivity of near infrared spectrum, and RED is red
The reflectivity of external spectrum.
3. winter wheat planting area extracting method according to claim 1, which is characterized in that described to use linear hybrid picture
First decomposition model is decomposed to obtain MODIS recognition result to the MODIS data, is specifically included:
The extraction of winter wheat image element information is carried out using linear spectral unmixing model, obtains MODIS winter wheat image element information;
According to the orthogonal vegetation index and time series of the MODIS data, typical feature time series indicatrix is obtained;
According to the typical feature time series indicatrix, endmember is selected by auxiliary data of the TM data, is utilized
Minimal noise separation conversion method classifies to the MODIS winter wheat image element information, obtains MODIS recognition result;It is described
MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat time series data and MODIS
Winter wheat pixel abundance figure, wherein MODIS winter wheat area and the non-winter wheat area the MODIS areWherein, when in typical feature time series indicatrix
Third wave band orthogonal vegetation index b3Vegetation index b orthogonal with second band2Difference be greater than the 0 and the 18th wave band
Orthogonal vegetation index b18Vegetation index b orthogonal with the 16th wave band16Difference less than 0 when, Pm=1 is MODIS winter wheat area;
Other situations, Pm=0 is the non-winter wheat area MODIS.
4. winter wheat planting area extracting method according to claim 1, which is characterized in that described small according to the TM winter
The area of wheat and the non-winter wheat area TM and MODIS winter wheat area and the non-winter wheat area MODIS carry out consistency analysis, determine one
Cause property and nonconforming demarcation threshold, specifically include:
Extract true winter wheat area in TM winter wheat area;
Obtain region recognition result identical with the true winter wheat area in the MODIS winter wheat area;
The Decomposition Accuracy of MODIS data mixing pixel is calculated according to the true winter wheat area and the region recognition result;
Picture in interval range where determining the abundance difference of winter wheat true value and MODIS recognition result according to the Decomposition Accuracy
Member accounts for the maximum ratio of whole collecting samples;
Demarcation threshold is determined according to the interval range.
5. winter wheat planting area extracting method according to claim 4, which is characterized in that described in the TM typical winter
The small area of wheat judges whether there is non-wheat mistake and enters, if so, according to time series using described in MODIS recognition result amendment
TM recognition result obtains the crop coverage measurement of the first winter wheat as a result, specifically including;
It judges whether there is non-winter wheat and is identified as winter wheat, if so, determining that there are non-wheat mistakes to enter;Utilize formulaCalculate the winter wheat pixel in TM typical case winter wheat area
Information, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRFor the winter wheat abundance after correction, T is point
Boundary's threshold value is 40%;
If it is not, then the winter wheat image element information in TM typical case winter wheat area is TM winter wheat abundance.
6. winter wheat planting area extracting method according to claim 4, which is characterized in that described in the TM atypia
Winter wheat area judges whether there is winter wheat mistake and goes out, if so, correcting institute using the MODIS recognition result according to time series
TM recognition result is stated, obtains the crop coverage measurement of the second winter wheat as a result, specifically including:
It judges whether there is part winter wheat and is identified as non-winter wheat, if so, determining that there are winter wheat mistakes to go out, and utilizes formulaCalculate the winter wheat pixel letter in the nonuniformity region
Breath, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRFor the winter wheat abundance after correction, T is boundary
Threshold value is 40%;
If it is not, then the winter wheat image element information in the TM atypia winter wheat area is TM winter wheat abundance.
7. a kind of winter wheat planting area extraction system characterized by comprising
Data capture unit, for obtaining TM data and MODIS data;The TM data are Medium resolution remotely sensed data through data
Obtained data are pre-processed, the MODIS data are the data that low resolution remote sensing images are obtained through data prediction;
TM recognition unit obtains TM identification knot for being classified using K-means non-supervised classification to the TM data
Fruit, the TM recognition result include TM winter wheat area, the non-winter wheat area TM and TM winter wheat pixel abundance figure;
MODIS recognition unit, for being decomposed to obtain to the MODIS data using linear spectral unmixing model
MODIS recognition result, the MODIS recognition result include MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat
Time series data and MODIS winter wheat pixel abundance figure;
Resampling unit obtains resampling TM data for carrying out resampling in default sample range;
Zoning unit, for determining TM typical case winter wheat area and TM atypia winter wheat area according to the resampling TM data;
Demarcation threshold determination unit, for according to TM winter wheat area and the non-winter wheat area TM and the MODIS winter wheat
Area and the non-winter wheat area MODIS carry out consistency analysis, determine consistency and nonconforming demarcation threshold;
Mistake enters amending unit, in TM typical case winter wheat area, judges whether there is non-wheat mistake and enters, if so, according to when
Between sequence using the MODIS recognition result correct the TM recognition result, obtain the crop coverage measurement knot of the first winter wheat
Fruit;
Mistake goes out amending unit, for judging whether there is winter wheat mistake and going out in the TM atypia winter wheat area, if so, according to
Time series corrects the TM recognition result using the MODIS recognition result, obtains the crop coverage measurement of the second winter wheat
As a result;
Combining unit, the growing surface for crop coverage measurement result and second winter wheat in conjunction with first winter wheat
Product extracts result and obtains the cultivated area of final winter wheat.
8. winter wheat planting area extraction system according to claim 7, which is characterized in that the data capture unit packet
It includes:
Geometric correction subelement, for being carried out using quadratic polynomial and bilinear interpolation method to the Medium resolution remotely sensed data
Geometric correction obtains TM data;
Cloud subelement is removed, for carrying out cloud removing to the low resolution remote sensing images, the MODIS image after obtaining cloud;
Orthogonal vegetation index computation subunit obtains MODIS for calculating orthogonal vegetation index using formula PVI=NIR-RED
Data, the MODIS data include the orthogonal vegetation index with time series;Wherein, PVI is orthogonal vegetation index, and NIR is
The reflectivity of near infrared spectrum, RED are the reflectivity of infrared spectroscopy.
9. winter wheat planting area extraction system according to claim 6, which is characterized in that the MODIS recognition unit
It specifically includes:
Image element information extracts subelement, for carrying out the extraction of winter wheat image element information using linear spectral unmixing model, obtains
To MODIS winter wheat image element information;
Indicatrix determines subelement, for the orthogonal vegetation index and time series according to the MODIS data, obtains typical case
Atural object time series indicatrix;
Classification subelement, for being selected by auxiliary data of the TM data according to the typical feature time series indicatrix
Endmember is selected, is classified using minimal noise separation conversion method to the MODIS winter wheat image element information, is obtained
MODIS recognition result;The MODIS recognition result includes MODIS winter wheat area, the non-winter wheat area MODIS, MODIS winter wheat
Time series data and MODIS winter wheat pixel abundance figure, wherein MODIS winter wheat area and the non-winter wheat of the MODIS
Qu WeiWherein, when typical feature time series feature is bent
The orthogonal vegetation index b of third wave band in line3Vegetation index b orthogonal with second band2Difference be greater than the 0 and the 18th wave
The orthogonal vegetation index b of section18Vegetation index b orthogonal with the 16th wave band16Difference less than 0 when, Pm=1 is small for the MODIS winter
The area of wheat;Other situations, Pm=0 is the non-winter wheat area MODIS.
10. winter wheat planting area extraction system according to claim 6, which is characterized in that the mistake enters amending unit
It is identified as winter wheat for judging whether there is non-winter wheat, if so, determining that there are non-wheat mistakes to enter;Utilize formulaCalculate the winter wheat pixel in TM typical case winter wheat area
Information, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRFor the winter wheat abundance after correction, T is point
Boundary's threshold value is 40%;If it is not, then the winter wheat image element information in TM typical case winter wheat area is TM winter wheat abundance;
The wrong amending unit out is identified as non-winter wheat for judging whether there is part winter wheat, if so, determination is deposited
Go out in winter wheat mistake, utilizes formulaIt calculates described non-uniform
The winter wheat image element information in property region, wherein PMODISFor MODIS winter wheat abundance, PTMFor TM winter wheat abundance, PRAfter correction
Winter wheat abundance, T is demarcation threshold, be 40%;If it is not, then the winter wheat image element information in the TM atypia winter wheat area is
TM winter wheat abundance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811219180.7A CN109242875B (en) | 2018-10-19 | 2018-10-19 | Winter wheat planting area extraction method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811219180.7A CN109242875B (en) | 2018-10-19 | 2018-10-19 | Winter wheat planting area extraction method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109242875A true CN109242875A (en) | 2019-01-18 |
CN109242875B CN109242875B (en) | 2020-08-18 |
Family
ID=65052439
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811219180.7A Active CN109242875B (en) | 2018-10-19 | 2018-10-19 | Winter wheat planting area extraction method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109242875B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110008905A (en) * | 2019-04-09 | 2019-07-12 | 安徽大学 | Regional scale stripe rust of wheat monitoring method based on the red side wave section of remote sensing image |
CN111860149A (en) * | 2020-06-11 | 2020-10-30 | 中科禾信遥感科技(苏州)有限公司 | Remote sensing identification method and device for overwintering rape and wheat |
CN115393732A (en) * | 2022-08-19 | 2022-11-25 | 南京林业大学 | Rice information extraction method based on MODIS time sequence image |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104615977A (en) * | 2015-01-26 | 2015-05-13 | 河南大学 | Winter wheat remote sensing recognition method capable of synthesizing key seasonal aspect characters and fuzzy classification technology |
CN104794424A (en) * | 2014-01-20 | 2015-07-22 | 北京天合数维科技有限公司 | Novel middle-low resolution remote sensing data combined cultivated land identification method |
CN105372672A (en) * | 2015-12-07 | 2016-03-02 | 广州地理研究所 | Time sequence data-based southern winter crop planting area extraction method |
-
2018
- 2018-10-19 CN CN201811219180.7A patent/CN109242875B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104794424A (en) * | 2014-01-20 | 2015-07-22 | 北京天合数维科技有限公司 | Novel middle-low resolution remote sensing data combined cultivated land identification method |
CN104615977A (en) * | 2015-01-26 | 2015-05-13 | 河南大学 | Winter wheat remote sensing recognition method capable of synthesizing key seasonal aspect characters and fuzzy classification technology |
CN105372672A (en) * | 2015-12-07 | 2016-03-02 | 广州地理研究所 | Time sequence data-based southern winter crop planting area extraction method |
Non-Patent Citations (4)
Title |
---|
YANSONG BAO 等: "Estimating winter wheat biomass based on LANDSAT TM and MODIS data", 《PROCEEDINGS OF SPIE》 * |
张军: "基于MODIS遥感数据的山东省济宁市冬小麦面积估算研究", 《中国优秀硕士学位论文全文数据库 农业科技辑》 * |
张霞 等: "基于 MODIS EVI 图像时间序列的冬小麦面积提取", 《农业工程学报》 * |
李新磊 等: "基于时序遥感数据的玉米种植面积提取方法研究基于时序遥感数据的玉米种植面积提取方法研究", 《农业网络信息》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110008905A (en) * | 2019-04-09 | 2019-07-12 | 安徽大学 | Regional scale stripe rust of wheat monitoring method based on the red side wave section of remote sensing image |
CN111860149A (en) * | 2020-06-11 | 2020-10-30 | 中科禾信遥感科技(苏州)有限公司 | Remote sensing identification method and device for overwintering rape and wheat |
CN111860149B (en) * | 2020-06-11 | 2024-01-12 | 中科禾信遥感科技(苏州)有限公司 | Remote sensing identification method and device for overwintering rape and wheat |
CN115393732A (en) * | 2022-08-19 | 2022-11-25 | 南京林业大学 | Rice information extraction method based on MODIS time sequence image |
Also Published As
Publication number | Publication date |
---|---|
CN109242875B (en) | 2020-08-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108921885B (en) | Method for jointly inverting forest aboveground biomass by integrating three types of data sources | |
CN111598019B (en) | Crop type and planting mode identification method based on multi-source remote sensing data | |
CN108764255B (en) | Method for extracting winter wheat planting information | |
CN106372592B (en) | A kind of winter wheat planting area calculation method based on winter wheat area index | |
CN106918816B (en) | Crop growth monitoring system and its data processing method and module | |
CN109685081B (en) | Combined change detection method for remote sensing extraction of abandoned land | |
CN110472184A (en) | A kind of cloudy misty rain area rice recognition methods based on Landsat remotely-sensed data | |
CN106126920B (en) | Crops disaster caused by hail disaster area remote sensing evaluation method | |
CN107527014A (en) | Crops planting area RS statistics scheme of sample survey design method at county level | |
CN108195767B (en) | Estuary wetland foreign species monitoring method | |
CN108932521B (en) | Deep learning-based crop classification method and system | |
CN109242875A (en) | A kind of winter wheat planting area extracting method and system | |
CN111340826A (en) | Single tree crown segmentation algorithm for aerial image based on superpixels and topological features | |
CN105809148A (en) | Crop drought recognition and risk evaluation method based on remote sensing time-space-spectrum fusion | |
CN113128401B (en) | Regional actual irrigation area monitoring method based on optical and radar remote sensing data | |
CN110414738A (en) | A kind of crop yield prediction technique and system | |
Wang et al. | Seasonality of leaf area index and photosynthetic capacity for better estimation of carbon and water fluxes in evergreen conifer forests | |
CN107688777A (en) | A kind of urban green space extracting method for cooperateing with multi-source Remote Sensing Images | |
Li et al. | Mapping rice cropping systems using Landsat-derived renormalized index of normalized difference vegetation index (RNDVI) in the Poyang Lake Region, China | |
CN110598514B (en) | Method for monitoring plot scale crop seeding area of land reclamation project area | |
Ouyang et al. | Assessment of canopy size using UAV-based point cloud analysis to detect the severity and spatial distribution of canopy decline | |
CN110245589A (en) | A kind of the Northwest's mulch information comprehensive analysis method based on high score remote sensing image | |
CN117218531B (en) | Sea-land ecological staggered zone mangrove plant overground carbon reserve estimation method | |
Li et al. | Urban vegetation mapping based on the hj-1 ndvi reconstrction | |
CN110245694A (en) | A kind of the Northwest's agricultural film information analysis method based on high score Remote Sensing Image Segmentation |
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 |