CN113642464B - Time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set - Google Patents
Time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set Download PDFInfo
- Publication number
- CN113642464B CN113642464B CN202110931179.2A CN202110931179A CN113642464B CN 113642464 B CN113642464 B CN 113642464B CN 202110931179 A CN202110931179 A CN 202110931179A CN 113642464 B CN113642464 B CN 113642464B
- Authority
- CN
- China
- Prior art keywords
- crop
- time sequence
- data
- ndvi
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set, comprising the following steps: step S1: acquiring remote sensing image data, land block data and crop sample data of a time sequence of a region to be detected; step S2: preprocessing the time series remote sensing image data; step S3: constructing an NDVI time sequence data set; step S4: respectively constructing NDVI time sequence data of different crop standards and NDVI time sequence data sets of land parcel units; step S5: constructing a TWDTW algorithm with non-equal-length time sequences, and obtaining minimum accumulated distance characteristics of similarity matching of different crops; step S6: calculating the climatic characteristics of different crop growing season lengths based on the NDVI time sequence data set of the land block units; step S7: based on the minimum accumulated distance characteristic and the growth season length characteristic, gaussian membership functions of different crops are constructed, and based on fuzzy set classification rules, fine classification of the crops on the land parcel scale is achieved. The invention realizes the fine classification of crops on the land parcel scale.
Description
Technical Field
The invention relates to the field of agricultural remote sensing, in particular to a time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set.
Background
The crop planting structure describes the type of regional crop planting and the spatial distribution information thereof, is important basic data for crop growth monitoring and yield estimation, and timely and accurately acquires the type of crops and the time-space variation information thereof, thereby having important significance for optimizing and adjusting the crop planting structure and reasonably configuring water and soil resources. The remote sensing technology becomes a main technical means for crop classification and space distribution information acquisition due to the advantages of wide coverage range, short observation period, strong behavior and the like.
When the remote sensing classification of various crops is carried out, overlapping and crossing exists among different crop spectrums, and the fine classification is difficult. The time sequence remote sensing image has certain advantages in representing the season phase and the change rule of the weather of crops, and is widely applied to multi-crop classification in complex areas. The dynamic time warping (Dynamic Time Warping, DTW) algorithm is a common method for time series similarity measurement, and has high flexibility, but is easy to cause serious mismatching phenomenon.
Disclosure of Invention
In view of the above, the present invention aims to provide a time-series remote sensing image crop classification method combining a dynamic time warping algorithm and a fuzzy set, which combines crop weather information, introduces a fuzzy set theory, and applies a time weighted TWDTW algorithm to crop classification research of the time-series remote sensing image to realize fine classification of crops on a plot scale.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set comprises the following steps:
step S1: acquiring remote sensing image data, land block data and crop sample data of a time sequence of a region to be detected;
step S2: preprocessing the time sequence remote sensing image data to obtain processed time sequence remote sensing data;
step S3: based on the time sequence remote sensing data, calculating the NDVI vegetation index classification characteristic to construct an NDVI time sequence data set;
step S4: based on the time sequence NDVI data set, combining crop sample data and land block data to respectively construct different crop standard NDVI time sequence data and NDVI time sequence data sets of land block units;
step S5: based on a dynamic time bending algorithm, constructing a TWDTW algorithm with non-equal-length time sequences, and obtaining minimum accumulated distance characteristics of similarity matching of different crops;
step S6: calculating the climatic characteristics of different crop growing season lengths based on the NDVI time sequence data set of the land block units;
step S7: based on the minimum accumulated distance characteristic and the growth season length characteristic, gaussian membership functions of different crops are constructed, and based on fuzzy set classification rules, fine classification of the crops on the land parcel scale is achieved.
Further, the preprocessing comprises radiation calibration, atmospheric correction, orthographic correction, image fusion, geographic registration and image clipping of the remote sensing image data.
Further, the step S4 is specifically that a characteristic value is extracted from the NDVI time sequence data by using crop sample data, and the standard NDVI time sequence data of each crop is obtained by averaging; and taking the land parcels as basic analysis units, solving an average value of the NDVI of all pixels in the land parcels by an average value filtering method, and assigning the average value to land parcels to obtain an NDVI time sequence data set of the land parcels.
Further, the TWDTW algorithm is:
let A be the pixel value of a certain position in the time series remote sensing image, A m =(a 1 ,a 2 ,a 3 ....,a m ) B is standard NDVI time sequence data of a certain type of crops to be identified, B n =(b 1 ,b 2 ,b 3 ....,b n ) M is larger than or equal to n, firstly, constructing a base distance matrix between A and B:
wherein D is ij Is a as i And b j The base distance between them, the Euclidean distance is used, namely:
according to the principle of minimum accumulated distance, a path with minimum accumulated distance is searched from a basic distance matrix according to a recursion method, the calculated accumulated distance at the tail end of the path is the minimum distance between time sequence data A and time sequence data B, and time sequences A and B are set in the matrixAt a cumulative distance of D ij The recursive calculation equation is:
when recursively searching the minimum accumulated distance path P, the boundary conditions in the calculation process are as follows:
introducing time weight, and calculating the following formula:
D′ ij =(1-λ)×D ij +λ×w ij (5)
wherein λ is a weighting coefficient of the time constraint factor, w ij Is a time weight factor, alpha is a gain factor, and the constant value is 0.1, delta t= |t i -t j I is the temporal distance between two matching points, β is typically taken as the intermediate node of the time series.
Further, the non-equal-length time sequence specifically includes: and combining the climate information of the crops to be classified, performing similarity matching with the time sequence NDVI data of the land block units in the same period by utilizing the standard NDVI time sequence data of each type of crop growing period according to the growing period lengths of different crops, and performing normalization processing on the similarity matching distance according to the growing period lengths of different crops to obtain the minimum accumulated distance characteristic of the crop similarity matching.
Further, the crop growth season length characteristic calculating method specifically comprises the following steps:
taking the time corresponding to the increase of the crop NDVI from the minimum value to 20% amplitude as a growth starting point, and starting photosynthesis when the crop is in the turning green or seedling stage;
the time corresponding to the reduction of the crop NDVI from the maximum value to 20% amplitude is the growth termination point, and at the moment, the crop is in the senescence or harvest period, and photosynthesis is slowed down or stopped;
the time length from the starting point to the ending point is the length of the growing season of the crops;
and (3) calculating and obtaining the crop growth season length characteristics based on the time sequence NDVI data of the land block units in the step S4.
Further, the fuzzy set classification rule specifically includes: construction of a Classification feature variable x i Is a gaussian membership function f (x i )∈{f(x 1 ),f(x 2 ),...,f(x n ) Setting the probability rule of the target land block object belonging to a certain crop as P i =max{f(x i ) And (3) comparing the probability of each crop to obtain the classification result of the land parcels.
Further, the gaussian membership function expression is as follows:
wherein x is i The method is characterized by classifying feature variables (such as the minimum accumulated distance feature of similarity measurement and the crop growth season length feature), mu and sigma are respectively two unknown parameter mean values and standard deviation of a Gaussian function, and specific values are determined according to sample values corresponding to different crops.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the fuzzy set theory is introduced by combining with the crop weather information, and the TWDTW algorithm is applied to crop classification research of time sequence remote sensing images, so that the fine classification of crops on the land parcel scale is realized.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the invention.
FIG. 2 is a timing diagram of a crop standard NDVI according to an embodiment of the present invention.
Fig. 3 is a plot-scale crop classification result diagram according to an embodiment of the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings and examples.
Referring to fig. 1, the present invention provides a time-series remote sensing image crop classification method combining TWDTW algorithm and fuzzy set, comprising the following steps:
step S1: acquiring time sequence remote sensing image data, land block data and crop sample data obtained by field investigation for crop classification;
step S2: carrying out pretreatment such as radiation calibration, atmospheric correction, orthographic correction, image fusion, geographic registration, image cutting and the like on remote sensing image data;
step S3: calculating the NDVI vegetation index classification characteristic based on the time sequence remote sensing data obtained by preprocessing in the step S2, and constructing an NDVI time sequence data set;
step S4: based on the time sequence NDVI data set, combining crop sample data and land block data to respectively construct crop standard NDVI time sequence data and an NDVI time sequence data set of a land block unit;
step S5: based on the step S4, a TWDTW crop classification method with non-equal-length time sequence is designed on the basis of a dynamic time warping algorithm (Dynamic Time Warping, DTW), and the minimum accumulated distance characteristic of crop similarity matching is obtained;
step S6: calculating the climatic characteristics of the crop growth season length based on the NDVI time sequence data set of the land block units;
step S7: and (3) integrating the minimum accumulated distance characteristic in the step S5 and the growth season length characteristic in the step S6, constructing Gaussian membership functions of different crops, and designing fuzzy set classification rules to realize the fine classification of the crops on the land parcel scale.
In this embodiment, step S1 specifically includes: obtaining a certain 2018 GF-1WFV time sequence remote sensing image, manually digitizing the 2018 GF-1PMS high-resolution remote sensing image serving as a reference base map to obtain land block vector data, and obtaining 560 main crop sample point data obtained by field investigation in 2018. Wherein, 28 sugar beet, 39 corn, 134 capsicum, 63 silage, 41 tomato, 191 wheat and 64 other crops.
In this embodiment, step S3 specifically includes: based on the preprocessed time sequence image, a 2018 NDVI time sequence data set is obtained through calculation.
In this embodiment, the step S4 specifically includes the following steps:
step S41: from crop sample data obtained by field investigation in 2018, 10 sample point crop training samples are randomly selected for each crop, crop standard NDVI time sequence data are constructed, and other sample points are verification samples; based on the NDVI time sequence image with 16m resolution, a grid with the size of 30m multiplied by 30m is constructed, the positions of 10 training sample data points of each crop are used as references, the corresponding grid is selected as planar sample data, values are extracted from the 15-scene NDVI time sequence image, the NDVI mean value of each planar sample is obtained through regional statistics, and then the NDVI of the 10 planar samples are averaged to obtain a crop standard NDVI time sequence curve (figure 2).
Step S42: taking a land parcel as a basic analysis unit, solving an average value of NDVI of all pixels in the land parcel by an average value filtering method, and assigning the average value to a land parcel object to obtain an NDVI time sequence data set of the land parcel in 2018;
in this embodiment, in step S5, the TWDTW algorithm is specifically implemented as follows:
let A be the pixel value of a certain position in 2018 time sequence NDVI image of a certain place, A m =(a 1 ,a 2 ,a 3 ....,a m ) B is standard NDVI time sequence data of crops (such as beet, corn, capsicum, silage, tomato and wheat) to be identified, B n =(b 1 ,b 2 ,b 3 ....,b n ) M is greater than or equal to n. First, a distance matrix between A and B is constructed:
wherein D is ij Is a as i And b j The base distance between them is generally the Euclidean distance, namely:
on the basis of the base distance, searching a path with the minimum accumulated distance from the base distance matrix according to the minimum accumulated distance principle and a recursion method, wherein the calculated accumulated distance at the tail end of the path is the minimum distance between the time sequence data A and the time sequence data B. Time series A and B are arranged inAt a cumulative distance of D ij The recursive calculation equation is:
when recursively searching the minimum accumulated distance path P, the boundary conditions which should be met in the calculation process are as follows:
time weight is introduced, and the calculation formula is as follows:
D′ ij =(1-λ)×D ij +λ×w ij (5)
wherein lambda is a weight coefficient of a time constraint factor, and the value in the experiment is 0.1, w ij Is a time weight factor, alpha is a gain factor, and the value in the experiment is 0.1, delta t= |t i -t j I is the temporal distance between two matching points, β is typically the time orderAnd the middle node is listed, and the value of beta is 100 in the experiment.
Preferably, the classification method of the non-equal-length time sequence TWDTW specifically comprises the following steps: according to different growth periods of crops, respectively setting classification intervals as follows: beet [116, 322] is 13-phase data, corn [129, 277] is 11-phase data, capsicum [129, 322] is 12-phase data, silage [129, 277] is 11-phase data, tomato [129, 240] is 9-phase data, and wheat [96, 191] is 7-phase data. And calculating the minimum accumulated distance characteristic of different crops by using the standard NDVI time sequence data of the crops and the NDVI time sequence data of the land parcel units in the corresponding growth interval.
In this embodiment, step S7 specifically includes: calculating and obtaining crop DTW minimum accumulated distance characteristic variable x by using 2018 non-isometric time sequence TWDTW optimal classification method 1 . NDVI time sequence data of a 2018 land parcel unit is utilized, and the NDVI of a key climatic period node in a crop growth curve is combined to calculate and obtain a crop growth season length characteristic variable x 2 . Sample values were extracted from the minimum cumulative distance and growing season length feature layers using 20 field survey sample point data for each crop, respectively, and the values of gaussian membership function statistical variables μ and σ were determined (table 1). Respectively constructing DTW minimum accumulated distance characteristic variable x based on 2018 crops 1 And crop growth season length characteristic variable x 2 Setting classification rule of target land block object belonging to a certain crop as P i =max{f(x 1 ),f(x 2 ) Crop classification results (fig. 3) of 2018 plot scale were obtained, and classification accuracy verification results are shown in table 2.
Table 1 statistical variable values of the crop gaussian membership function
TABLE 2 classification accuracy
In the example, the total precision OA of the non-equal-length time sequence classification method is 0.86, the kappa coefficient is 0.82, and the TWDTW classification method of the fuzzy set is introduced, so that the method has certain application potential in the fine classification of crops in the land parcel scale.
The foregoing description is only of the preferred embodiments of the invention, and all changes and modifications that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims (5)
1. The time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set is characterized by comprising the following steps:
step S1: acquiring remote sensing image data, land block data and crop sample data of a time sequence of a region to be detected;
step S2: preprocessing the time sequence remote sensing image data to obtain processed time sequence remote sensing data;
step S3: based on the time sequence remote sensing data, calculating the NDVI vegetation index classification characteristic to construct an NDVI time sequence data set;
step S4: based on the time sequence NDVI data set, combining crop sample data and land block data to respectively construct different crop standard NDVI time sequence data and NDVI time sequence data sets of land block units;
step S5: based on a dynamic time bending algorithm, constructing a time weighted dynamic algorithm with non-equal time sequences, and obtaining the minimum accumulated distance characteristics of similarity matching of different crops;
step S6: calculating the climatic characteristics of different crop growing season lengths based on the NDVI time sequence data set of the land block units;
step S7: constructing Gaussian membership functions of different crops based on the minimum accumulated distance features and the growth season length features, and realizing fine classification of the crops on the scale of land parcels based on fuzzy set classification rules;
the TWDTW algorithm is as follows:
let A be the pixel value of a certain position in the time series remote sensing image, A m =(a 1 ,a 2 ,a 3 ....,a m ) B is standard NDVI time sequence data of a certain type of crops to be identified, B n =(b 1 ,b 2 ,b 3 ....,b n ) M is larger than or equal to n, firstly, constructing a base distance matrix between A and B:
in the method, in the process of the invention,is a as i And b j The base distance between them, the Euclidean distance is used, namely:
according to the principle of minimum accumulated distance, a path with minimum accumulated distance is searched from a basic distance matrix according to a recursion method, the calculated accumulated distance at the tail end of the path is the minimum distance between time sequence data A and time sequence data B, and time sequences A and B are set in the matrixAt a cumulative distance of D ij The recursive calculation equation is:
when recursively searching the minimum accumulated distance path P, the boundary conditions in the calculation process are as follows:
introducing time weight, and calculating the following formula:
D' ij =(1-λ)×D ij +λ×w ij (5)
wherein w is ij Is a time weight factor, alpha is a gain factor, and the constant value is 0.1, delta t= |t i -t j The I is the time distance between two matching points, and beta is generally the middle node of the time sequence;
the non-equal-length time sequence specifically comprises the following steps: combining the climate information of the crops to be classified, performing similarity matching with the time sequence NDVI data of the land block units in the same period by utilizing the standard NDVI time sequence data of each type of crop growing period according to the growing period lengths of different crops, and performing normalization processing on the similarity matching distance according to the growing period lengths of different crops to obtain the minimum accumulated distance characteristic of crop similarity matching;
the fuzzy set classification rule specifically comprises the following steps: construction of a Classification feature variable x i Is a gaussian membership function f (x i )∈{f(x 1 ),f(x 2 ),…,f(x n ) Setting the probability rule of the target land block object belonging to a certain crop as P i =max{f(x i ) And (3) comparing the probability of each crop to obtain the classification result of the land parcels.
2. The time-series remote sensing image crop classification method combining the TWDTW algorithm and the fuzzy set according to claim 1, wherein the preprocessing comprises performing radiation calibration, atmospheric correction, orthographic correction, image fusion, geographic registration and image cropping on the remote sensing image data.
3. The method for classifying the time-series remote sensing image crops by combining the TWDTW algorithm and the fuzzy set according to claim 1 is characterized in that the step S4 is specifically that the characteristic value is extracted from the NDVI time-series data by using crop sample data, and the standard NDVI time-series data of each crop is obtained by averaging; and taking the land parcels as basic analysis units, solving an average value of the NDVI of all pixels in the land parcels by an average value filtering method, and assigning the average value to land parcels to obtain an NDVI time sequence data set of the land parcels.
4. The time-series remote sensing image crop classification method combining TWDTW algorithm and fuzzy set according to claim 1, wherein the crop growth season length feature calculation method is specifically as follows:
taking the time corresponding to the increase of the crop NDVI from the minimum value to 20% amplitude as a growth starting point, and starting photosynthesis when the crop is in the turning green or seedling stage;
the time corresponding to the reduction of the crop NDVI from the maximum value to 20% amplitude is the growth termination point, and at the moment, the crop is in the senescence or harvest period, and photosynthesis is slowed down or stopped;
the time length from the starting point to the ending point is the length of the growing season of the crops;
and (3) calculating and obtaining the crop growth season length characteristics based on the time sequence NDVI data of the land block units in the step S4.
5. The time-series remote sensing image crop classification method combining TWDTW algorithm and fuzzy set according to claim 2, wherein the Gaussian membership function expression is as follows:
wherein x is i Is a classification characteristic variable, and mu and sigma are respectively the mean value and standard deviation of two unknown parameters of a Gaussian function.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110931179.2A CN113642464B (en) | 2021-08-13 | 2021-08-13 | Time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110931179.2A CN113642464B (en) | 2021-08-13 | 2021-08-13 | Time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113642464A CN113642464A (en) | 2021-11-12 |
CN113642464B true CN113642464B (en) | 2023-09-15 |
Family
ID=78421547
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110931179.2A Active CN113642464B (en) | 2021-08-13 | 2021-08-13 | Time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113642464B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114926565B (en) * | 2022-05-24 | 2024-04-12 | 中国科学院东北地理与农业生态研究所 | Soybean drafting method based on green degree and water comprehensive index |
CN114782825B (en) * | 2022-06-17 | 2022-09-16 | 中国农业科学院农业资源与农业区划研究所 | Crop identification method and device based on incomplete remote sensing data and electronic equipment |
CN115620152B (en) * | 2022-12-16 | 2023-03-14 | 中化现代农业有限公司 | Rice identification method and device, electronic equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110119717A (en) * | 2019-05-15 | 2019-08-13 | 中国科学院遥感与数字地球研究所 | A kind of Crop classification method based on multi-temporal NDVI and LST |
CN111598019A (en) * | 2020-05-19 | 2020-08-28 | 华中农业大学 | Crop type and planting mode identification method based on multi-source remote sensing data |
CN111950361A (en) * | 2020-07-07 | 2020-11-17 | 内蒙古农业大学 | Beet identification method based on single-time-sequence NDVI |
CN112395914A (en) * | 2019-08-15 | 2021-02-23 | 中国科学院遥感与数字地球研究所 | Method for identifying land parcel crops by fusing remote sensing image time sequence and textural features |
WO2021098471A1 (en) * | 2019-11-19 | 2021-05-27 | 浙江大学 | Wide-range crop phenology extraction method based on morphological modeling method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11328506B2 (en) * | 2019-12-26 | 2022-05-10 | Ping An Technology (Shenzhen) Co., Ltd. | Crop identification method and computing device |
-
2021
- 2021-08-13 CN CN202110931179.2A patent/CN113642464B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110119717A (en) * | 2019-05-15 | 2019-08-13 | 中国科学院遥感与数字地球研究所 | A kind of Crop classification method based on multi-temporal NDVI and LST |
CN112395914A (en) * | 2019-08-15 | 2021-02-23 | 中国科学院遥感与数字地球研究所 | Method for identifying land parcel crops by fusing remote sensing image time sequence and textural features |
WO2021098471A1 (en) * | 2019-11-19 | 2021-05-27 | 浙江大学 | Wide-range crop phenology extraction method based on morphological modeling method |
CN111598019A (en) * | 2020-05-19 | 2020-08-28 | 华中农业大学 | Crop type and planting mode identification method based on multi-source remote sensing data |
CN111950361A (en) * | 2020-07-07 | 2020-11-17 | 内蒙古农业大学 | Beet identification method based on single-time-sequence NDVI |
Non-Patent Citations (2)
Title |
---|
基于GF-1/WFVNDVI时间序列数据的作物分类;杨闫君;占玉林;田庆久;顾行发;余涛;王磊;;农业工程学报(第24期);全文 * |
基于MODIS时间序列及物候特征的农作物分类;平跃鹏;臧淑英;;自然资源学报(第03期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113642464A (en) | 2021-11-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113642464B (en) | Time sequence remote sensing image crop classification method combining TWDTW algorithm and fuzzy set | |
CN108921885B (en) | Method for jointly inverting forest aboveground biomass by integrating three types of data sources | |
Fang et al. | Corn‐yield estimation through assimilation of remotely sensed data into the CSM‐CERES‐Maize model | |
Johnson | Temporal stability of an NDVI‐LAI relationship in a Napa Valley vineyard | |
US7167800B2 (en) | Method of optimizing remote sensing operation timing | |
CN106780091A (en) | Agricultural disaster information remote sensing extracting method based on vegetation index time space statistical nature | |
CN110363246B (en) | Fusion method of vegetation index NDVI with high space-time resolution | |
CN103971199B (en) | A kind of remote sensing ranking method of a wide range of Grain Growth Situation | |
CN109063660B (en) | Crop identification method based on multispectral satellite image | |
Skakun et al. | Winter wheat yield assessment using Landsat 8 and Sentinel-2 data | |
CN111275567B (en) | SENP cotton yield estimation method based on unmanned aerial vehicle image and estimation model construction method | |
CN113609899B (en) | Remote sensing time sequence analysis-based tilling land information positioning display system | |
CN114882359B (en) | Soybean planting area extraction method and system based on vegetation index time series spectrum characteristics | |
CN112836725A (en) | Weak supervision LSTM recurrent neural network rice field identification method based on time sequence remote sensing data | |
CN116188465B (en) | Crop growth state detection method based on image processing technology | |
CN111723711A (en) | Plianes and object-oriented mulching film information extraction method and system | |
CN116824384A (en) | Soybean identification method based on standard curve | |
CN114926743A (en) | Crop classification method and system based on dynamic time window | |
Ohana-Levi et al. | Time-series clustering of remote sensing retrievals for defining management zones in a vineyard | |
Borgogno-Mondino et al. | Describing the spatio-temporal variability of vines and soil by satellite-based spectral indices: A case study in Apulia (South Italy) | |
CN110390287A (en) | A kind of crop maturity phase prediction technique based on satellite remote sensing | |
CN116579521B (en) | Yield prediction time window determining method, device, equipment and readable storage medium | |
Berd et al. | The use of NDVI algorithm in predicting the productivity of rice fields of Talang District of Solok Regency | |
WO2023055897A1 (en) | Methods and systems for use in processing images related to crops | |
CN115661674A (en) | Crop irrigation information extraction method based on multisource satellite soil humidity data |
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 |