CN115512237A - Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data - Google Patents
Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data Download PDFInfo
- Publication number
- CN115512237A CN115512237A CN202211253356.7A CN202211253356A CN115512237A CN 115512237 A CN115512237 A CN 115512237A CN 202211253356 A CN202211253356 A CN 202211253356A CN 115512237 A CN115512237 A CN 115512237A
- Authority
- CN
- China
- Prior art keywords
- ndvi
- winter wheat
- data
- remote sensing
- research area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Image Processing (AREA)
Abstract
The invention relates to an Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data, which comprises the following steps of: acquiring Sentinel-2A image data including visible light and near red wave bands of a target area, and preprocessing a complete remote sensing image of a research area; obtaining a multi-temporal vegetation index NDVI; obtaining maximum value synthetic NDVI data; identifying winter wheat in a research area by adopting a decision tree to obtain a primary classification result graph; obtaining an image of a winter wheat planting area in a research area; and selecting sample points according to the phenological characteristics and the spectral characteristics of the winter wheat, and performing precision verification on the results after the winter wheat is classified to generate a precision analysis report. According to the method, a key time phase for identifying the winter wheat is screened out according to the phenological characteristics of the winter wheat, and the winter wheat is identified by combining multi-scene image data with a vegetation index NDVI; the invention carries out processing such as cutting, inlaying and correcting on the original data, improves the precision of the image data and further improves the precision of identifying the winter wheat.
Description
Technical Field
The invention relates to the technical field of remote sensing image identification, in particular to an Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data.
Background
The method has great significance for guaranteeing national grain production safety by timely and accurately acquiring crop planting distribution and planting area information. The method can quickly and accurately acquire the distribution of ground crops based on the analysis of satellite images of earth observation satellites, and is one of the main means for acquiring the information of the wide-range crop planting distribution. The Anhui province is a big winter wheat planting province in China, and the annual planting area exceeds four million mu.
At present, in the method for identifying the planting distribution of winter wheat in Anhui province, crops are classified mainly based on single-time phase or full-time sequence images, the adopted method has the problems that key phenological information of the crops is lost, the growth characteristics of the crops are ignored, the classification effect of a single experience threshold or a combined threshold is greatly influenced by human subjectivity, partial key period data is shielded by cloud layers and the like, the types of the crops are complex and various, and obvious spectral overlap exists among different crops, so that the phenomena of 'wrong classification and missing classification' are easy to occur when the single-time phase remote sensing image data are used for classifying the crops, the ideal classification precision is difficult to achieve, and the planting distribution and area extraction result of the winter wheat in Anhui province is not ideal.
The phenological characteristics of the plants enable the plants to show different spectral characteristics at different growth and development stages, and the multi-temporal remote sensing data can describe the spectral difference of the same crop at different growth and development stages, so that the time series remote sensing image has great application potential in the aspects of crop planting information extraction, yield prediction, agricultural condition monitoring and the like. The vegetation index can effectively reflect the phenological differences and spectral characteristics of different vegetation types, and improves the classification precision to the crop level, so that the vegetation index is widely applied to the extraction and research of crop planting information. The normalized vegetation index NDVI is one of the most widely applied spectral parameters in crop monitoring, and has important practical guiding significance in crop production and management.
Disclosure of Invention
The invention aims to provide the Anhui winter wheat planting area extraction method based on the multi-temporal satellite remote sensing data, which has higher resolution, can more accurately identify winter wheat plots, reduces the dependence on manual experience and professional knowledge, and improves the classification precision.
In order to realize the purpose, the invention adopts the following technical scheme: a method for extracting the planting area of Anhui winter wheat based on multi-temporal satellite remote sensing data comprises the following steps in sequence:
(1) Acquiring Sentinel-2A image data including visible light and near red wave bands of a target area, and preprocessing a complete remote sensing image of a research area;
(2) Calculating remote sensing of the preprocessed image research area according to an NDVI calculation formula to obtain multi-temporal vegetation indexes NDVI;
(3) Taking the characteristics of key growth period data required by extraction of winter wheat into consideration, carrying out maximum synthesis operation on data of several adjacent periods capable of representing growth period characteristics to obtain maximum synthesis NDVI data;
(4) According to the growth characteristics and the phenological characteristics of wheat, identifying winter wheat in a research area by adopting a decision tree to obtain a primary classification result graph;
(5) Masking the preliminary classification result graph by using a farmland vector file of a research area, obtaining an image of a winter wheat planting area in a research area;
(6) And selecting sample points according to the phenological characteristics and the spectral characteristics of the winter wheat, and performing precision verification on the results after the winter wheat is classified to generate a precision analysis report.
In the step (1), the preprocessing of the complete remote sensing image of the research area refers to: the method comprises the steps of obtaining remote sensing images of a research area in different periods, carrying out cloud cover screening to obtain first data, carrying out radiometric calibration, atmospheric correction, geometric correction and resampling on the first data to obtain second data, and cutting and inlaying the obtained second data according to the research area to obtain time sequence image data of the research area.
In the step (2), the multi-temporal vegetation index NDVI is calculated by the following formula:
NDVI＝(B8-B4)/(B8+B4)
in the formula, B8 represents a Sentinel-2 near infrared band, and B4 represents a Sentinel-2 red band.
In the step (3), the key growth period required by the extraction of the winter wheat comprises a sowing period, a seedling period, a green turning period, a jointing period and a period of maximum biomass of the previous crop; the maximum synthesis of the two-stage data is realized through spectral calculation, and the formula is as follows:
B4ⅹ(B4 ge B8)+B8ⅹ(B8 gt B4)
in the formula, B8 represents a Sentinel-2 near infrared band, B4 represents a Sentinel-2 red band, ge represents more than or equal to, and gt represents more than or equal to;
when the data is more than two-phase, the maximum value is obtained by cycling in the ArcGis software.
The step (4) specifically comprises the following steps:
(4a) Finding out expressions of each node of the decision tree:
Node1＝(NDVI 9 gt NDVI 11 )and(NDVI 11 gt NDVI 10 )and(NDVI 9 gt 0.48)
Node2＝(NDVI 4-2 gt 0.2)and(NDVI 2 gt 0.29)
in the formula: NDVI 2 NDVI is synthesized in the middle of 2 months at the maximum; NDVI 4 NDVI is synthesized in 4 middle of the month; NDVI 9 NDVI is synthesized in the middle ten days of the month to the maximum extent; NDVI 10 NDVI is synthesized in 10 middle of the month at the maximum; NDVI 11 NDVI is synthesized in 11 middle of the month to the maximum; gt represents greater than;
(4b) Executing the decision tree by adopting ENVI software:
the color of two non-winter wheat of Class1 and Class2 was defined as White and the Class3 preliminary winter wheat color was defined as Green, respectively.
The step (5) specifically comprises the following steps: and (3) for the result graph of the preliminary classification of the research area, masking part of the cultivated land by using the existing cultivated land vector file of the research area and a masking tool in ArcGIS software to obtain an image of the winter wheat planting area of the research area.
The step (6) specifically comprises the following steps: obtaining sample points with correct classification and sample points with wrong classification by using the selected wheat sample points and the wheat planting area image of the research area, drawing a confusion matrix, and calculating two indexes of total classification precision and a kappa coefficient, so as to obtain the classification precision, wherein the larger the total classification precision and the kappa coefficient is, the higher the classification precision is, and the calculation formula of the total classification precision is as follows:
in the formula: p c For overall classification accuracy, n is the total row and column number of the confusion matrix; p kk Diagonal elements of the confusion matrix; p is the total number of samples;
the kappa coefficient is calculated as:
in the formula: k is a kappa coefficient, and N is the number of total pixels; r is the total row and column number of the confusion matrix; x ii Diagonal elements of the confusion matrix; x i+ Is the sum of the rows of the confusion matrix; x +i Is the sum of the columns of the confusion matrix.
According to the technical scheme, the beneficial effects of the invention are as follows: firstly, according to the phenological characteristics of winter wheat, screening out a key time phase of winter wheat identification, identifying the winter wheat by combining multi-scene image data with a vegetation index NDVI, calculating the NDVI, converting the multi-waveband data into an independent image waveband for displaying vegetation distribution, wherein a higher NDVI value indicates that more green vegetation is contained, and the method can well reflect the phenological information of the vegetation; secondly, the original data are cut, embedded, corrected and the like, so that the precision of the image data is improved, and the precision of identifying the winter wheat is further improved; thirdly, the method extracts the planting area of the winter wheat based on the Sentinel-2A data, adopts the Sentinel-2A data, has higher resolution compared with the traditional MODIS and Landsat data, and more accurately identifies the winter wheat land parcel by combining the phenological characteristics of the winter wheat; fourthly, the method classifies the crops based on the decision tree, supports the independent learning of different characteristics of the crops, fully excavates the association between characteristic data, reduces the dependence on human experience and professional knowledge, and improves the classification precision.
Drawings
FIG. 1 is a schematic diagram of a decision tree;
FIG. 2 is a schematic view of sample positioning;
FIG. 3 is a flow chart of the method of the present invention.
Detailed Description
As shown in fig. 3, a method for extracting the planting area of Anhui winter wheat based on multi-temporal satellite remote sensing data comprises the following steps in sequence:
(1) Acquiring Sentinel-2A image data including visible light and near red wave bands of a target area, and preprocessing a complete remote sensing image of a research area;
(2) Calculating remote sensing of the preprocessed image research area according to an NDVI calculation formula to obtain multi-temporal vegetation indexes NDVI;
(3) Taking the characteristics of key growth period data required by extraction of winter wheat into consideration, carrying out maximum synthesis operation on data of several adjacent periods capable of representing growth period characteristics to obtain maximum synthesis NDVI data;
(4) According to the growth characteristics and the phenological characteristics of wheat, identifying winter wheat in a research area by adopting a decision tree to obtain a primary classification result graph;
(5) Masking the preliminary classification result image by using a farmland vector file of the research area to obtain an image of a winter wheat planting area of the research area;
(6) And selecting sample points according to the phenological characteristics and the spectral characteristics of the winter wheat, and performing precision verification on the results after the winter wheat is classified to generate a precision analysis report.
In the step (1), the preprocessing of the complete remote sensing image of the research area includes: the method comprises the steps of obtaining remote sensing images of a research area in different periods, carrying out cloud amount screening to obtain first data, carrying out radiometric calibration, atmospheric correction, geometric correction and resampling on the first data to obtain second data, and cutting and inlaying the obtained second data according to the research area to obtain time sequence image data of the research area.
In the step (2), the calculation formula of the vegetation index NDVI of the multiple time phases is as follows:
NDVI＝(B8-B4)/(B8+B4)
in the formula, B8 represents a Sentinel-2 near infrared band, and B4 represents a Sentinel-2 red band.
In the step (3), the key growth periods required by the winter wheat comprise a sowing period, a seedling period, a green turning period, a jointing period and a period of maximum biomass of the previous crop; the maximum synthesis of the two-stage data is realized through spectral calculation, and the formula is as follows:
B4ⅹ(B4 ge B8)+B8ⅹ(B8 gt B4)
in the formula, B8 represents a Sentinel-2 near infrared band, B4 represents a Sentinel-2 red band, ge represents more than or equal to, and gt represents more than or equal to;
when the data is more than two-phase, the maximum value is obtained by cycling in the ArcGis software. In other words, in actual practice, the maximum value synthesis NDVI data is directly implemented by the ArcGis software.
The step (4) specifically comprises the following steps:
as shown in fig. 1, (4 a) finding the expressions of the nodes of the decision tree:
Node1＝(NDVI 9 gt NDVI 11 )and(NDVI 11 gt NDVI 10 )and(NDVI 9 gt 0.48)
Node2＝(NDVI 4-2 gt 0.2)and(NDVI 2 gt 0.29)
in the formula: NDVI 2 NDVI is synthesized in the middle of 2 months at the maximum; NDVI 4 NDVI is synthesized in 4 middle of the month; NDVI 9 NDVI is synthesized in the middle ten days of the month to the maximum extent; NDVI 10 NDVI is synthesized in 10 middle of the month at the maximum; NDVI 11 NDVI is synthesized in 11 middle of the month to the maximum; gt represents greater than;
(4b) Executing the decision tree by adopting ENVI software:
the color of two non-winter wheat of Class1 and Class2 was defined as White and the Class3 preliminary winter wheat color was defined as Green, respectively.
In actual operation, a decision tree is adopted on ENVI software to identify winter wheat in a research area to obtain a primary classification result graph.
The step (5) specifically comprises the following steps: and (3) for the result graph of the preliminary classification of the research area, masking part of the cultivated land by using the existing cultivated land vector file of the research area and a masking tool in ArcGIS software to obtain an image of the winter wheat planting area of the research area.
The step (6) specifically comprises the following steps: obtaining sample points with correct classification and sample points with wrong classification by using the selected wheat sample points and the wheat planting area image of the research area, drawing a confusion matrix, and calculating two indexes of total classification precision and a kappa coefficient, so as to obtain the classification precision, wherein the larger the total classification precision and the kappa coefficient is, the higher the classification precision is, and the calculation formula of the total classification precision is as follows:
in the formula: p c For the overall classification accuracy, n is the total row and column number of the confusion matrix; p kk Diagonal elements of the confusion matrix; p is the total number of samples;
the kappa coefficient is calculated as:
in the formula: k is a kappa coefficient, and N is the number of total pixels; r is the total row and column number of the confusion matrix; x ii Diagonal elements of the confusion matrix; x i+ Is the sum of the rows of the confusion matrix; x +i Is the sum of the columns of the confusion matrix.
The invention is further described below with reference to fig. 1 to 3.
Example one
The multi-temporal satellite-based remote sensing data adopted by the invention can select corresponding earth observation satellite data, such as MODIS data (American space agency, with the highest spatial resolution of 250 m), sentinel No. 2 data (European space agency, with the highest spatial resolution of 10 m) and WorldView satellite data (digital global company, with the highest spatial resolution of 0.5 m), according to different scale requirements or data acquisition costs. The embodiment of the invention takes the sentinel number 2 data as an example. Sentinel 2 is a high resolution multispectral imaging satellite, carries a multispectral imager, is used for land monitoring, can provide images such as vegetation, soil and water coverage, inland waterway and coastal region, divide into two satellites of 2A and 2B. The height of the sentinel No. 2 satellite is 786km, the sentinel covers 13 spectral bands, and the breadth reaches 290 kilometers. The ground resolution is respectively 10m, 20m and 60m, the revisit period of one satellite is 10 days, the two satellites are complementary, and the revisit period is 5 days. In this embodiment, the red bands (Band 4, B4) and the near infrared bands (Band 8, B8) of the data of the sentinel 2A are used to calculate the NDVI of the land, and table 1 shows the parameters of the bands of the sentinel 2A.
TABLE 1 parameters of each wave band of the sentinel 2A
When the remote sensing image is preprocessed, cloud screening refers to screening data which contains cloud occlusion in the obtained image data, images with the cloud amount smaller than 10% are reserved, and the image data after cloud screening, namely the first data, is obtained.
Since the resolution of the B4 and B8 bands in the sentinel No. 2 data is the same, no resampling step is required using the sentinel No. 2 data. If the resolution of other satellite data bands is different, each band needs to be resampled to the same resolution.
The reason for cutting and inlaying the second data is that the obtained original data can not cover the complete research area, and several images need to be inlaid together, and then the redundant part is cut off to finally obtain the complete research area.
And (3) acquiring observation data of several key growth periods of winter wheat growth by considering the phenological characteristics of winter wheat planting in Anhui province, performing maximum synthesis operation on multi-period data which can represent the adjacent period of growth period characteristics, and calculating to obtain multiple growth period maximum value synthesized NDVI data.
Firstly, acquiring key growth period data required for monitoring the area of winter wheat from (n-1) th year to nth year, wherein the key growth period data comprises NDVI (winter wheat seedling stage) in middle of 2 months (winter wheat seedling stage), NDVI (winter wheat seedling stage) in middle of 4 months (winter wheat growth stage), NDVI (crop biomass maximum stage) in middle of 9 months (previous crop biomass maximum stage), NDVI in middle of 10 months (winter wheat seeding stage) and NDVI in middle of 11 months (winter wheat seedling stage); table 2 is a synthetic description of the critical growth period data required to monitor winter wheat area from (n-1) th to n th years.
TABLE 2
Decision trees are a widely used classification algorithm. Compared with Bayesian algorithms, decision trees have the advantage that no domain knowledge or parameter settings are required for the construction process. The decision tree is a tree structure (which may be a binary tree or a non-binary tree). Each non-leaf node represents a test on a feature attribute, each branch represents the output of the feature attribute over a range of values, and each leaf node stores a category. The process of using the decision tree to make a decision is to start from the root node, test the corresponding characteristic attributes in the items to be classified, select an output branch according to the value of the characteristic attributes until the leaf node is reached, and take the category stored by the leaf node as a decision result.
The decision tree node expression is shown in table 3. The color of both non-winter wheat of Class1 and Class2 was defined as White and the Class3 preliminary winter wheat color as Green.
TABLE 3 decision Tree node definition
The remote sensing processing software adopted by the invention is ENVI, and the following introduces the specific operation steps of decision tree classification:
(1) Starting ENVI and opening required data;
(2) Establishing a Decision Tree, wherein in the Toolbox on the right, classification/Decision Tree/New Decision Tree;
(3) Double-click, editing the decision tree on the popped interface, clicking the uppermost node, and inputting a name and an expression;
(4) Clicking OK, popping up a variable selection template;
(5) Clicking NDVI on the popped interface, popping a panel, selecting a file, clicking OK, and finishing variable selection;
(6) According to the method, the construction of the whole decision tree is completed;
(7) Clicking in the blank, selecting and executing the decision tree, setting parameters on a pop-up interface, clicking OK, and starting calculation.
When winter wheat sample points are selected according to the phenological characteristics and the spectral characteristics of wheat, rape is a crop with a growth period similar to that of wheat, in the middle and last 3 months, the rape is in a flowering period, the plot is yellow, the winter wheat is in a rising period, the plot is dark green, in the middle and last April, the rape is thanks, the plot is green, the wheat is in a growth period of vigorous vigour, the plot is dark green, in the last 5 months to last 6 months, the rape and the wheat are in a harvesting period, and the plot is in a bare soil state. At least 50 wheat samples are selected according to the characteristics, and the samples are far away from non-cultivated areas such as villages, roads, water bodies, industrial and mining areas and the like. To ensure that the sample (250 mx 250 m) is a pure pixel on the remote sensing image, a plot of winter wheat all in the range of (750 mx 750 m) is selected, the center of the GPS sample is used for positioning, the positioning accuracy is within 3 meters, the sample interval is at least 10 meters, and a sample longitude and latitude information text file is established, as shown in FIG. 2.
And performing precision verification on the classification result by adopting a confusion matrix, wherein two index parameters of total classification precision and Kappa are mainly adopted. The classification precision is required to reach more than 90%, the classification precision is influenced by environmental factors, and the growth vigor of winter wheat at the same time and the same place and different time intervals is different, so when the extraction precision is lower than 90%, the threshold value of the expression in the decision tree needs to be properly adjusted according to the actual condition of a research area until the expression reaches the standard.
In conclusion, according to the phenological characteristics of winter wheat, the key time phase of winter wheat identification is screened out, the winter wheat identification is carried out by combining multi-scene image data with a vegetation index NDVI, the NDVI can be calculated to convert the multi-waveband data into an independent image waveband for displaying vegetation distribution, a higher NDVI value indicates that more green vegetation is contained, and the method can well reflect the phenological information of the vegetation; the method performs cutting, embedding, correction and other processing on the original data, improves the precision of the image data, and further improves the precision of identifying the winter wheat; according to the method, the winter wheat planting area is extracted based on the Sentinel-2A data, the Sentinel-2A data is adopted, the resolution is higher compared with the traditional MODIS and Landsat data, and the winter wheat plots are accurately identified by combining the phenological characteristics of the winter wheat; the invention classifies crops based on the decision tree, supports the independent learning of different characteristics of the crops, fully excavates the association between characteristic data, reduces the dependence on human experience and professional knowledge, and improves the classification precision.
Claims (7)
1. A method for extracting Anhui winter wheat planting area based on multi-temporal satellite remote sensing data is characterized by comprising the following steps: the method comprises the following steps in sequence:
(1) Acquiring Sentinel-2A image data including visible light and near red wave bands of a target area, and preprocessing a complete remote sensing image of a research area;
(2) Calculating remote sensing of the preprocessed image research area according to an NDVI calculation formula to obtain multi-temporal vegetation indexes NDVI;
(3) Taking the characteristics of key growth period data required by extraction of winter wheat into consideration, carrying out maximum synthesis operation on data of several adjacent periods capable of representing growth period characteristics to obtain maximum synthesis NDVI data;
(4) According to the growth characteristics and the phenological characteristics of wheat, identifying winter wheat in a research area by adopting a decision tree to obtain a primary classification result graph;
(5) Masking the preliminary classification result image by using a farmland vector file of the research area to obtain an image of a winter wheat planting area of the research area;
(6) And selecting sample points according to the phenological characteristics and the spectral characteristics of the winter wheat, and performing precision verification on the results after the winter wheat is classified to generate a precision analysis report.
2. The Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data according to claim 1, characterized in that: in the step (1), the preprocessing of the complete remote sensing image of the research area includes: the method comprises the steps of obtaining remote sensing images of a research area in different periods, carrying out cloud cover screening to obtain first data, carrying out radiometric calibration, atmospheric correction, geometric correction and resampling on the first data to obtain second data, and cutting and inlaying the obtained second data according to the research area to obtain time sequence image data of the research area.
3. The Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data according to claim 1, characterized in that: in the step (2), the calculation formula of the vegetation index NDVI of the multiple time phases is as follows:
NDVI＝(B8-B4)/(B8+B4)
in the formula, B8 represents a Sentinel-2 near infrared band, and B4 represents a Sentinel-2 red band.
4. The Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data according to claim 1, characterized in that: in the step (3), the key growth period required by the extraction of the winter wheat comprises a sowing period, a seedling period, a green turning period, a jointing period and a period of maximum biomass of the previous crop; the maximum synthesis of the two-stage data is realized through spectral calculation, and the formula is as follows:
B4ⅹ(B4 ge B8)+B8ⅹ(B8 gt B4)
in the formula, B8 represents a Sentinel-2 near infrared band, B4 represents a Sentinel-2 red band, ge represents more than or equal to, and gt represents more than or equal to;
when the data is more than two-phase, the maximum value is obtained by cycling in the ArcGis software.
5. The Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data according to claim 1, characterized in that: the step (4) specifically comprises the following steps:
(4a) Finding out expressions of each node of the decision tree:
Node1＝(NDVI 9 gt NDVI 11 )and(NDVI 11 gt NDVI 10 )and(NDVI 9 gt 0.48)
Node2＝(NDVI 4-2 gt 0.2)and(NDVI 2 gt 0.29)
in the formula: NDVI 2 NDVI in middle of 2 months is synthesized at the maximum; NDVI 4 NDVI is synthesized in 4 middle of the month at the maximum; NDVI 9 NDVI is synthesized in the middle ten days of the month to the maximum extent; NDVI 10 NDVI is synthesized in 10 middle of the month at the maximum; NDVI 11 NDVI is synthesized in 11 middle of the month to the maximum; gt represents greater than;
(4b) Executing the decision tree by adopting ENVI software:
the color of two non-winter wheat of Class1 and Class2 was defined as White and the Class3 preliminary winter wheat color was defined as Green, respectively.
6. The Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data according to claim 1, characterized in that: the step (5) specifically comprises the following steps: and (3) for the result graph of the preliminary classification of the research area, masking part of the cultivated land by using the existing cultivated land vector file of the research area and a masking tool in ArcGIS software to obtain an image of the winter wheat planting area of the research area.
7. The Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data according to claim 1, characterized in that: the step (6) specifically comprises the following steps: obtaining sample points with correct classification and sample points with wrong classification by using the selected wheat sample points and the wheat planting area image of the research area, drawing a confusion matrix, and calculating two indexes of total classification precision and a kappa coefficient, so as to obtain the classification precision, wherein the larger the total classification precision and the kappa coefficient is, the higher the classification precision is, and the calculation formula of the total classification precision is as follows:
in the formula: p c For overall classification accuracy, n is the total row and column number of the confusion matrix; p kk Diagonal elements of the confusion matrix; p is the total number of samples;
the kappa coefficient is calculated as:
in the formula: k is kappa coefficient, and N is the number of total pixels; r is the total row and column number of the confusion matrix; x ii Diagonal elements of the confusion matrix; x i+ Is the sum of the rows of the confusion matrix; x +i Is the sum of the columns of the confusion matrix.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211242032 | 2022-10-11 | ||
| CN2022112420323 | 2022-10-11 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115512237A true CN115512237A (en) | 2022-12-23 |
Family
ID=84509879
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211253356.7A Pending CN115512237A (en) | 2022-10-11 | 2022-10-13 | Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115512237A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116485784A (en) * | 2023-06-12 | 2023-07-25 | 天地信息网络研究院(安徽)有限公司 | Winter idle field remote sensing extraction method based on time sequence NDVI |
| CN117830860A (en) * | 2024-03-06 | 2024-04-05 | 江苏省基础地理信息中心 | Remote sensing automatic extraction method of winter wheat planting structure |
-
2022
- 2022-10-13 CN CN202211253356.7A patent/CN115512237A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116485784A (en) * | 2023-06-12 | 2023-07-25 | 天地信息网络研究院(安徽)有限公司 | Winter idle field remote sensing extraction method based on time sequence NDVI |
| CN116485784B (en) * | 2023-06-12 | 2023-09-12 | 天地信息网络研究院(安徽)有限公司 | Winter idle field remote sensing extraction method based on time sequence NDVI |
| CN117830860A (en) * | 2024-03-06 | 2024-04-05 | 江苏省基础地理信息中心 | Remote sensing automatic extraction method of winter wheat planting structure |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Blickensdörfer et al. | Mapping of crop types and crop sequences with combined time series of Sentinel-1, Sentinel-2 and Landsat 8 data for Germany | |
| Waldhoff et al. | Multi-Data Approach for remote sensing-based regional crop rotation mapping: A case study for the Rur catchment, Germany | |
| Dong et al. | Tracking the dynamics of paddy rice planting area in 1986–2010 through time series Landsat images and phenology-based algorithms | |
| George et al. | Forest tree species discrimination in western Himalaya using EO-1 Hyperion | |
| CN115512237A (en) | Anhui winter wheat planting area extraction method based on multi-temporal satellite remote sensing data | |
| CN109685081B (en) | Combined change detection method for remote sensing extraction of abandoned land | |
| CN108458978B (en) | Sensitive waveband and waveband combination optimal tree species multispectral remote sensing identification method | |
| CN107527014A (en) | Crops planting area RS statistics scheme of sample survey design method at county level | |
| CN114821362B (en) | Multi-source data-based rice planting area extraction method | |
| CN106918816A (en) | crop growth monitoring system and its data processing method and module | |
| CN113205014B (en) | Time sequence data farmland extraction method based on image sharpening | |
| US20220392215A1 (en) | System and Method for Mapping Land Cover Types with Landsat, Sentinel-1, and Sentinel-2 Images | |
| CN111738066A (en) | Gridding rice sheath blight habitat evaluation method integrating multi-source remote sensing information | |
| CN111462223A (en) | Method for identifying soybean and corn planting area in Jianghuai region based on Sentinel-2 image | |
| CN116543316B (en) | Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image | |
| Zeng et al. | A phenology-based vegetation index classification (PVC) algorithm for coastal salt marshes using Landsat 8 images | |
| Nguyen et al. | Medium-resolution multispectral satellite imagery in precision agriculture: Mapping precision canola (Brassica napus L.) yield using Sentinel-2 time series | |
| Liu et al. | Mapping the complex crop rotation systems in Southern China considering cropping intensity, crop diversity, and their seasonal dynamics | |
| CN111079846A (en) | Apple identification method based on time series high-resolution remote sensing data | |
| CN112329733A (en) | Winter wheat growth monitoring and analyzing method based on GEE cloud platform | |
| Duan et al. | Mapping saffron fields and their ages with Sentinel-2 time series in north-east Iran | |
| Yang et al. | Monitoring winter wheat in ShanDong province using Sentinel data and Google Earth Engine platform | |
| GASTELLU-ETCHEGORRY et al. | Computer-assisted land cover mapping with SPOT in Indonesia | |
| Singla et al. | Extraction of Crop Information from Reconstructed Landsat Data in Himalayan Foothills Region | |
| Garcia-Torres et al. | Semi-automatic normalization of multitemporal remote images based on vegetative pseudo-invariant features |
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 |