CN111695533B - Remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year - Google Patents
Remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year Download PDFInfo
- Publication number
- CN111695533B CN111695533B CN202010548468.XA CN202010548468A CN111695533B CN 111695533 B CN111695533 B CN 111695533B CN 202010548468 A CN202010548468 A CN 202010548468A CN 111695533 B CN111695533 B CN 111695533B
- Authority
- CN
- China
- Prior art keywords
- year
- farmland
- winter wheat
- farmlands
- time sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
- G09B29/003—Maps
- G09B29/006—Representation of non-cartographic information on maps, e.g. population distribution, wind direction, radiation levels, air and sea routes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Ecology (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year, which comprises the following steps: 1) constructing smooth MODIS NDVI time sequence data; 2) obtaining year-by-year farmland and non-farmland drawings based on a machine learning algorithm and the reconstructed MODIS NDVI time sequence data; 3) transforming the reconstructed MODIS NDVI time sequence data to obtain a daily-scale NDVI time sequence, marking farmlands in a year-by-year farmland and non-farmland classification result diagram to obtain a year-by-year multiple-cropping monitoring drawing; 4) and obtaining an annual winter wheat and summer corn planting area map based on a machine learning algorithm, the reconstructed MODIS NDVI time sequence data and an annual multiple planting monitoring map. The year-by-year continuous mapping result of the remote sensing mapping method can reflect the planting time and space change conditions of winter wheat and summer corn in North China plain, also considers the actual condition of farmland use continuity, and the data product can directly provide data support and decision reference for industry departments.
Description
Technical Field
The invention belongs to the technical field of remote sensing image processing, and particularly relates to a remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year.
Background
North China plain is one of the important main food production areas in China, and bears the burden of food supply and safety in China, and winter wheat-summer corn rotation is the main planting mode in the area. The method has the advantages that the planting condition of winter wheat and summer corn, particularly the spatial distribution of sowing and the change trend condition of the sowing over the years are known in detail and accurately, the method has important practical significance for the optimized layout, large-scale planting, standardized production and industrialized operation of regional grain production, and meanwhile, the method has important reference value for management departments to make relevant land use policies and grain subsidy policy plans in time.
The existing traditional method generally refers to the annual grain seeding area statistical data provided by each level of statistical departments. Although the method has the advantages that the planting condition of the winter wheat and the summer corn can be mastered in counties and above scales and total amount, the method also has two defects that the workload is large, the statistical period is long, and the planting condition of the winter wheat and the summer corn in each province or the whole north China can be known only after the statistics departments at all levels report and summarize; and secondly, the accurate spatial position information of winter wheat and summer corn seeding is difficult to master. Scholars also put forward a mapping method for winter wheat seeding monitoring based on remote sensing images, but the methods generally only aim at mapping for a single year planting area, but lack consideration on farmland use continuity, and therefore the method is not good in automatic monitoring mapping year by year. The existing remote sensing image monitoring and mapping method has the outstanding problems that the spatial position of the sowing area of winter wheat changes greatly year by year, and a year-by-year planting area map which is consistent with the actual situation is difficult to obtain.
Disclosure of Invention
In order to overcome the defects of the traditional statistical method and the conventional method, the invention aims to provide the remote sensing mapping method for automatically monitoring the planting area of the winter wheat and the summer corn year by year, the method can reflect the planting time and space change conditions of the winter wheat and the summer corn in North China plain according to the year by year continuous mapping result, the actual condition of the use continuity of the farmland is also considered, and the data product can directly provide data support and decision reference for the industry department.
The invention relates to a remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year, which is characterized by comprising the following steps:
1) constructing smooth MODIS NDVI time sequence data;
2) classifying the remote sensing images based on a machine learning algorithm and the reconstructed MODIS NDVI time sequence data to obtain year-by-year farmland and non-farmland drawings;
3) transforming the reconstructed MODIS NDVI time sequence data to obtain a daily-scale NDVI time sequence, marking farmlands in a year-by-year farmland and non-farmland classification result diagram to obtain a year-by-year multiple-cropping monitoring drawing;
4) based on a machine learning algorithm, reconstructed MODIS NDVI time sequence data and a year-by-year multiple-species monitoring map, remote sensing images are classified to obtain a year-by-year winter wheat and summer corn planting area map.
Further, the step 1) uses a Savitzky-Golay filter to construct the smoothed MODIS NDVI time series data.
Furthermore, the specific method for obtaining the year-by-year farmland and non-farmland drawing in the step 2) is as follows:
2.1) classifying farmlands and non-farmlands in the remote sensing images of the current calculation year based on a machine learning algorithm, selecting the remote sensing images of winter wheat and summer corn in the key growth period, selecting three machine learning algorithms of a neural network, a decision tree and a support vector machine according to high-resolution image training sample data of the farmlands and the non-farmlands, respectively constructing and classifying classifiers for the farmlands and the non-farmlands, wherein the classification judgment rule is as follows: at least two machines are judged to be farmlands on the same region/grid which is judged to be the farmlands, the utilization map of the farmland and non-farmland subsoil of the current calculation year is obtained,
acquiring the correlation and significance of normalized vegetation index NDVI change on grid scale of adjacent years based on reconstructed MODIS NDVI time sequence data, judging whether the farmland has land use type change in the use process based on the correlation and significance of the normalized vegetation index NDVI change, and judging whether the farmland has land use type change in the use process of the farmland and non-farmland subsoil use type change areas in the current calculation year by adopting a classifier constructed by a neural network again, wherein the significance p is less than 0.05, and directly inheriting the adjacent background year type to obtain a classification result graph of the farmland and non-farmland year.
Furthermore, the specific method for obtaining the multiple-cropping-year monitoring map in step 3) is to perform six-degree polynomial fitting on the reconstructed MODIS NDVI time series data to obtain a daily-scale NDVI time series, and according to the daily-scale NDVI time series, multiple use cases of the farmland in the multiple-cropping-year farmland and non-farmland classification result graphs can be obtained, wherein the multiple use cases include a single-cropping or double-cropping planting mode, the time and the corresponding maximum value of the peak value of the crop in the single-cropping mode, the time and the corresponding maximum value of the peak value of the crop in the first cropping and second cropping modes in the double-cropping mode, and whether a abandoned land exists.
Furthermore, the specific method for obtaining the year-by-year planting area map of the winter wheat and the summer corn in the step 3) comprises the following steps:
4.1) classifying the remote sensing images of the current calculation year based on a neural network method and high-resolution image training samples of winter wheat and summer corn to obtain a mapping of the planting area of the winter wheat and the summer corn of the current calculation year, wherein the high-resolution image training samples are obtained by field survey of the unmanned aerial vehicle, Google Earth and the like,
4.2) respectively calculating the correlation and the significance of the normalized vegetation index NDVI in two growth periods of winter wheat and summer corn, acquiring the correlation and the significance of the change of the normalized vegetation index NDVI on a grid scale of adjacent years based on reconstructed MODIS NDVI time sequence data, judging whether the planted crops change or not according to the correlation and the significance of the change of the normalized vegetation index (NDVI), constructing a classifier based on a neural network method, reclassifying the changed areas in the drawing of the planted areas of the winter wheat and the summer corn of the current calculation year, directly inheriting the types of the planted areas of the adjacent background years by the significantly correlated areas with the significance p of less than 0.05, and obtaining the drawing of the planted areas of the winter wheat and the summer corn one by one.
Further, in the present invention,
the method has the beneficial effects that 1) the method analyzes the northern China plain replanting condition based on year-by-year continuous winter wheat and summer corn planting area maps, and comprises various replanting mode time and space pattern changes (such as single-season, double-season and abandoned land change conditions; as another example, winter wheat-summer maize rotation, winter wheat-other crop rotation or other crop-summer maize rotation); 2) the method builds a Savitzky-Golay filter to reconstruct a high-quality time sequence MODIS NDVI data set; 3) adopting various machine learning algorithms (such as the most widely used neural network, decision tree and support vector machine) to classify and optimize farmland and non-farmland; 4) performing continuous farmland drawing based on the machine learning algorithm with the best performance and the correlation and significance of the reconstructed high-quality MODIS NDVI time sequence data on adjacent years on the grid scale; 5) and (4) performing multiple species monitoring mapping based on reconstructed high-quality MODIS NDVI time sequence data and an annual farmland map. The method comprises a single-season or double-season planting mode, the time of the peak value of the crop in the single-season planting mode and the corresponding maximum value, the time of the peak value of the crop in the first season and the second season in the double-season planting mode and the corresponding maximum value, the existence of the abandoned land possibility and other information, and can obtain more accurate planting area drawings of the winter wheat and the summer corn; 6) intermediate data of a winter wheat and summer corn planting area drawing is obtained based on a best-performing machine learning algorithm and a high-quality training data set (sample plot data obtained based on unmanned aerial vehicles, field surveys and high-resolution image data), and a continuous winter wheat and summer corn planting area drawing is obtained based on a best-performing machine learning algorithm, adjacent year correlation and significance of high-quality MODIS NDVI time sequence data on a grid scale and a year-by-year farmland multiple monitoring drawing.
Compared with the traditional statistical method and the conventional method, the remote sensing mapping method for automatically monitoring the planting area of the winter wheat and the summer corn year by year, which is developed by the invention, can reflect the planting time and space change condition of the winter wheat and the summer corn in North China plain according to the year-by-year continuous mapping result, also takes the actual condition of the use continuity of the farmland into consideration, and the inverted data product can directly provide data support and decision reference for the industry department.
Drawings
FIG. 1 is a flow chart of the remote sensing mapping method for automatically monitoring the planting area of winter wheat and summer corn year by year according to the invention;
FIG. 2 is a schematic diagram of the construction of high quality MODIS NDVI time series data;
FIG. 3 is a remote sensing classification chart of Farmland and non-Farmland in North China plain (2019);
fig. 4 is a space distribution diagram of winter wheat and summer corn planting in north China plain (2019).
Detailed Description
The following structural description and the accompanying drawings further describe the specific technical scheme of the invention.
The original data adopted by the method is MODIS data in 2019 and MODIS data obtained in field investigation of unmanned aerial vehicles, Google Earth and the like are used as sample data.
As shown in the attached figure 1, the remote sensing mapping method for automatically monitoring the planting areas of winter wheat and summer corn year by year specifically comprises the following steps:
1) smooth MODIS NDVI time series data were constructed.
Since the original MODIS NDVI time series data are generally affected by the noise due to the influence of cloud and atmospheric conditions, the invention first constructs a Savitzky-Golay filter to perform minimum second smoothing on the original MODIS NDVI time series data, as shown in fig. 2. Taking the data processing in the ith year as an example, the specific construction steps are as follows: to reduce the noise of the early and late year data, the last three-phase image of the last year (if any) and the first three-phase image of the next year (if any) are also introduced into the smoothing filter algorithm; renumbering all images used for smoothing input in the (i-1) th year, the (i +1) th year (the specific coding rule used in the invention is shown in figure 2), so that the Savitzky-Golay filter can read data conveniently; after the data preparation is finished, carrying out data smoothing processing by using a Savitzky-Golay filter; when the filter data is output, only the data of the ith year is output (the data is not output in the (i-1) th year and the (i +1) th year) and the original code is recovered, and finally the reconstructed high-quality MODIS NDVI time sequence data of the ith year is obtained. The method can be repeatedly used to obtain high-quality MODIS NDVI time sequence data of all years. In the invention, four splines are selected according to parameter configuration conditions in the Savitzky-Golay filter, the size of a sliding window is set to be 9(2 x 4+1), and the iteration times are 6.
2) And classifying the remote sensing images based on the machine learning algorithm and the reconstructed MODIS NDVI time sequence data to obtain year-by-year farmland and non-farmland charting.
2.1) classifying farmlands and non-farmlands in the remote sensing images Of the current calculation year based on a machine learning algorithm, wherein one practicable method is to select the remote sensing images Of the key growth periods Of winter wheat and summer corn, which relate to DOY (day Of Yeast) 64-289 and comprise 15 scenes every year. High-resolution image training sample data of farmlands and non-farmlands in remote sensing image areas are obtained based on Google Earth, and classifiers are constructed and classified for the farmlands and the non-farmlands in the remote sensing image areas respectively by selecting and using three machine learning algorithms of a neural network, a decision tree and a support vector machine which are most extensive. The three machine learning algorithms have advantages, so that the farmland results are optimized by integrating the three classification results. The specific optimization method is that at least two types of machine learning are satisfied, the farmland is judged in the same area/grid, and the area is judged as the farmland. The method is used for obtaining the utilization map of the farmland and non-farmland subsoil of the current calculation year.
2.2) based on the reconstructed MODIS NDVI time sequence data (DOY64-289), the correlation and the significance of the change of the normalized vegetation index (NDVI) on the grid scale of adjacent years can be obtained, and whether the land utilization type change occurs in the farmland in the using process is judged according to the correlation and the significance of the change of the normalized vegetation index (NDVI). The NDVI time series correlation R for the current year x and the adjacent year y is calculated as follows:
wherein R is i,j Representing the correlation coefficient, x, at spatial position (i, j) i,j And y i,j NDVI time series data set representing the current year x and the adjacent year y at spatial position (i, j), n being the number of time series. The corresponding significance (i.e., p-value) of the correlation coefficient was calculated by student's t-test (p is used in the present invention)<0.05 as statistical criteria by significance test). Through the analysis of the calculation process of the farmland and non-farmland background land utilization maps of the current calculation year, the overall precision and kappa coefficient of the classification result of the neural network-based method are generally better than those of other two algorithms, so that the neural network method is used in the year-by-year continuous farmland and non-farmland drawing process. For farmland and non-farmland background land profit of current calculation yearClassifying and judging the regions with changed land use types by using a classifier constructed by a neural network again in the graph, and obtaining the regions with significant correlation (significance p)<0.05) directly inheriting the type of the adjacent background year to obtain a classification result graph of the farmland and the non-farmland year by year, as shown in the attached figure 3;
3) and transforming the reconstructed MODIS NDVI time sequence data to obtain a daily-scale NDVI time sequence, marking farmlands in a year-by-year farmland and non-farmland classification result diagram, and obtaining a year-by-year multiple-species monitoring chart.
And performing six-order polynomial fitting on the reconstructed MODIS NDVI time sequence data to obtain a daily scale NDVI time sequence, and thus obtaining multiple use conditions of the farmland in a year-by-year farmland and non-farmland classification result diagram, wherein the multiple use conditions comprise information such as a single-season or double-season planting mode, the time of occurrence of a crop peak value and a corresponding maximum value in the single-season planting mode, the time of occurrence of a crop peak value and a corresponding maximum value in the first season and the second season in the double-season planting mode, possibility of abandoned land existence and the like. For example, if a computation region/grid exhibits only one NDVI peak, this indicates a single seasonal value;
4) based on a machine learning algorithm, reconstructed MODIS NDVI time sequence data and a year-by-year multiple-species monitoring map, remote sensing images are classified to obtain a year-by-year winter wheat and summer corn planting area map.
4.1) classifying the remote sensing images of the current calculation year based on a machine learning algorithm and high-resolution image training samples of winter wheat and summer corn to obtain a planting area map of the winter wheat and the summer corn of the current calculation year. The high-resolution image training samples can be obtained through field survey of the unmanned aerial vehicle, Google Earth and the like, a classifier is constructed by using a neural network method in the embodiment, the winter wheat and the summer corn are classified and extracted, the planting area is mapped, the winter wheat and summer corn planting area map of the training sample in the corresponding year is obtained, the current calculation year of the embodiment is 2019, and the map is shown in the attached figure 4.
4.2) the method of this step is somewhat similar to the farmland classification method of step 2). However, the method of this step requires the calculation of normalized vegetation index (NDVI) correlation and significance for both the winter wheat and summer maize growing periods, respectively, the winter wheat selection period being DOY64-DOY177 annually (total 8 scenes) and the summer maize selection period being DOY177-DOY289 annually (total 8 scenes). And (3) judging whether the planted crops change according to the correlation and the significance of the change of the normalized vegetation index (NDVI) (the correlation and the significance judging method is detailed in a formula 1), constructing a classifier based on a neural network method, and classifying and judging changed areas in the drawing of the planted winter wheat and summer corn areas of the current calculation year again, wherein the significantly related areas (the significance p is less than 0.05) directly inherit the types of the crops of the adjacent background year to obtain the year-by-year planted winter wheat and summer corn areas drawing.
The method of the invention analyzes the North China plain replanting situation based on the year-by-year continuous winter wheat and summer corn planting area map, including the time and space pattern changes of various replanting modes. The method can acquire the planting area maps (2000-2019) of the North China plain winter wheat and summer corn in all time periods of MODIS data, analyze the change conditions of single season, double season and abandoned land, and simultaneously analyze the change conditions of multiple types (such as winter wheat-summer corn rotation, winter wheat-other crop rotation or other crop-summer corn rotation), thereby providing data support and decision support for industry departments.
Although the present invention has been described in terms of the preferred embodiment, it is not intended that the invention be limited to the embodiment. Any equivalent changes or modifications made without departing from the spirit and scope of the present invention also belong to the protection scope of the present invention. The scope of the invention should therefore be determined with reference to the appended claims.
Claims (2)
1. A remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year is characterized by comprising the following steps:
1) constructing reconstructed MODIS NDVI time sequence data;
2) the remote sensing images are classified based on a machine learning algorithm and the reconstructed MODIS NDVI time sequence data to obtain year-by-year farmland and non-farmland drawings, and the specific method for obtaining the year-by-year farmland and non-farmland drawings comprises the following steps:
2.1) classifying farmlands and non-farmlands in the remote sensing images of the current calculation year based on a machine learning algorithm, selecting the remote sensing images of winter wheat and summer corn in the key growth period, selecting three machine learning algorithms of a neural network, a decision tree and a support vector machine according to high-resolution image training sample data of the farmlands and the non-farmlands, respectively constructing and classifying classifiers for the farmlands and the non-farmlands, wherein the classification judgment rule is as follows: at least two machines are judged to be farmlands on the same region/grid which is judged to be the farmlands, the utilization map of the farmland and non-farmland subsoil of the current calculation year is obtained,
2.2) acquiring correlation and significance of normalized vegetation index NDVI change on grid scale of adjacent years based on reconstructed MODIS NDVI time sequence data, judging whether land use type change occurs in the farmland in the using process based on the correlation and significance of the normalized vegetation index NDVI change, and judging whether the land use type change occurs in the farmland and non-farmland subsoil utilization areas in the current calculation year by adopting a classifier constructed by a neural network, wherein the significant correlation areas with significance p less than 0.05 directly inherit the adjacent background year types to obtain a classification result graph of the farmland and non-farmland year by year;
3) transforming the reconstructed MODIS NDVI time sequence data to obtain an NDVI time sequence of a daily scale, marking farmlands in a classification result diagram of year-by-year farmlands and non-farmlands to obtain a year-by-year multiple cropping monitoring drawing, wherein the specific method for obtaining the year-by-year multiple cropping monitoring drawing is to perform six-degree polynomial fitting on the reconstructed MODIS NDVI time sequence data to obtain the day-scale NDVI time sequence, and according to the day-scale NDVI time sequence, multiple cropping use conditions of the farmlands in the classification result diagram of year-by-year farmlands and non-farmlands can be obtained, wherein the multiple cropping use conditions comprise a single-season or double-season planting mode, the occurrence time of a crop peak in the single-season planting mode and a corresponding maximum value, and the occurrence time of the crop peak in the first season and the second season in the double-season planting mode and a corresponding maximum value and whether abandoned land exists or not;
4) classifying remote sensing images based on a machine learning algorithm, reconstructed MODIS NDVI time sequence data and a year-by-year multiple-species monitoring map to obtain a year-by-year winter wheat and summer corn planting area map, wherein the specific method for obtaining the year-by-year winter wheat and summer corn planting area map comprises the following steps:
4.1) classifying the remote sensing images of the current calculation year based on a neural network method and high-resolution image training samples of winter wheat and summer corn to obtain a mapping of the planting area of the winter wheat and the summer corn of the current calculation year, wherein the high-resolution image training samples are obtained by field survey of the unmanned aerial vehicle, Google Earth and the like,
4.2) respectively calculating the correlation and the significance of the normalized vegetation index NDVI in two growth periods of winter wheat and summer corn, acquiring the correlation and the significance of the change of the normalized vegetation index NDVI on a grid scale of adjacent years based on reconstructed MODIS NDVI time sequence data, judging whether the planted crops change or not according to the correlation and the significance of the change of the normalized vegetation index (NDVI), constructing a classifier based on a neural network method, reclassifying the changed areas in the drawing of the planted areas of the winter wheat and the summer corn of the current calculation year, directly inheriting the types of the planted areas of the adjacent background years by the significantly correlated areas with the significance p of less than 0.05, and obtaining the drawing of the planted areas of the winter wheat and the summer corn one by one.
2. The method for remotely sensing and mapping the planting areas of winter wheat and summer corn on the basis of automatic year-by-year monitoring as claimed in claim 1, wherein the reconstructed MODIS NDVI time series data in the step 1) are obtained by smoothing through a Savitzky-Golay filter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010548468.XA CN111695533B (en) | 2020-06-16 | 2020-06-16 | Remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010548468.XA CN111695533B (en) | 2020-06-16 | 2020-06-16 | Remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111695533A CN111695533A (en) | 2020-09-22 |
CN111695533B true CN111695533B (en) | 2022-08-12 |
Family
ID=72481379
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010548468.XA Active CN111695533B (en) | 2020-06-16 | 2020-06-16 | Remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111695533B (en) |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102592181A (en) * | 2011-12-30 | 2012-07-18 | 中国农业科学院农业资源与农业区划研究所 | Method for optimizing spatial distribution of statistical data about crop planting area |
CN104951772B (en) * | 2015-07-10 | 2017-12-29 | 中国科学院遥感与数字地球研究所 | A kind of winter wheat extracting method based on NDVI time-serial positions integration |
CN108955620B (en) * | 2018-02-13 | 2019-08-16 | 中国科学院遥感与数字地球研究所 | A kind of method and system of farmland irrigated area area Remotely sensed acquisition |
CN109685108B (en) * | 2018-11-23 | 2022-05-03 | 电子科技大学 | Method for generating high-space-time resolution NDVI long-time sequence |
CN110751094B (en) * | 2019-10-21 | 2020-08-28 | 北京师范大学 | Crop yield estimation method based on GEE comprehensive remote sensing image and deep learning method |
CN110909679B (en) * | 2019-11-22 | 2020-09-18 | 中国气象科学研究院 | Remote sensing identification method and system for fallow crop rotation information of winter wheat historical planting area |
-
2020
- 2020-06-16 CN CN202010548468.XA patent/CN111695533B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN111695533A (en) | 2020-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11275197B2 (en) | Forecasting national crop yield during the growing season | |
CN112889089B (en) | Machine learning techniques for identifying clouds and cloud shadows in satellite imagery | |
US11361184B2 (en) | Modeling trends in crop yields | |
CN111598019B (en) | Crop type and planting mode identification method based on multi-source remote sensing data | |
WO2017196756A1 (en) | Statistical blending of weather data sets | |
CA2981473C (en) | Forecasting national crop yield during the growing season | |
BR112019015401A2 (en) | PLANTATION YIELD ESTIMATE USING AGRONOMIC NEURAL NETWORK | |
AU2017211495A1 (en) | System and method for earth observation and analysis | |
EP3873187A1 (en) | Automated sample collection and tracking system | |
CN115829162B (en) | Crop yield prediction method, device, electronic equipment and medium | |
US11610272B1 (en) | Predicting crop yield with a crop prediction engine | |
CN109711102A (en) | A kind of crop casualty loss fast evaluation method | |
Peters et al. | Synergy of very high resolution optical and radar data for object-based olive grove mapping | |
US11704581B1 (en) | Determining crop-yield drivers with multi-dimensional response surfaces | |
US20220196877A1 (en) | Forecasting national crop yield during the growing season | |
BR112021007721A2 (en) | Machine learning improved plant disease infection detection | |
CN116543316B (en) | Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image | |
WO2023108213A1 (en) | Methods and systems for classifying and benchmarking irrigation performance | |
Patenaude et al. | Integrating remote sensing datasets into ecological modelling: a Bayesian approach | |
US11715024B1 (en) | Estimating soil chemistry at different crop field locations | |
Pelta et al. | Forecasting seasonal plot-specific crop coefficient (Kc) protocol for processing tomato using remote sensing, meteorology, and artificial intelligence | |
CN111695533B (en) | Remote sensing mapping method for automatically monitoring planting areas of winter wheat and summer corn year by year | |
CN114782835B (en) | Crop lodging area proportion detection method and device | |
US11704576B1 (en) | Identifying ground types from interpolated covariates | |
WO2023055897A1 (en) | Methods and systems for use in processing images related to crops |
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 |