CN114118679A - Crop yield per unit and growth evaluation method based on time sequence remote sensing data - Google Patents
Crop yield per unit and growth evaluation method based on time sequence remote sensing data Download PDFInfo
- Publication number
- CN114118679A CN114118679A CN202111200212.0A CN202111200212A CN114118679A CN 114118679 A CN114118679 A CN 114118679A CN 202111200212 A CN202111200212 A CN 202111200212A CN 114118679 A CN114118679 A CN 114118679A
- Authority
- CN
- China
- Prior art keywords
- yield
- time
- vegetation index
- crop
- remote sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000012010 growth Effects 0.000 title claims abstract description 95
- 238000011156 evaluation Methods 0.000 title claims abstract description 47
- 230000001186 cumulative effect Effects 0.000 claims abstract description 57
- 238000000034 method Methods 0.000 claims abstract description 40
- 238000012544 monitoring process Methods 0.000 claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 238000011160 research Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 7
- 238000004519 manufacturing process Methods 0.000 claims description 7
- 238000010899 nucleation Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000000611 regression analysis Methods 0.000 claims description 6
- 238000005520 cutting process Methods 0.000 claims description 4
- 238000003908 quality control method Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 5
- 244000068988 Glycine max Species 0.000 description 31
- 235000010469 Glycine max Nutrition 0.000 description 31
- 230000000694 effects Effects 0.000 description 10
- 238000012360 testing method Methods 0.000 description 10
- 238000004458 analytical method Methods 0.000 description 9
- 238000011161 development Methods 0.000 description 9
- 230000018109 developmental process Effects 0.000 description 9
- 238000012417 linear regression Methods 0.000 description 6
- 238000009331 sowing Methods 0.000 description 4
- 238000009825 accumulation Methods 0.000 description 3
- 238000011835 investigation Methods 0.000 description 3
- 239000002028 Biomass Substances 0.000 description 2
- 238000010220 Pearson correlation analysis Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 230000009666 qualitative growth Effects 0.000 description 2
- 238000002310 reflectometry Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 244000061176 Nicotiana tabacum Species 0.000 description 1
- 235000002637 Nicotiana tabacum Nutrition 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 241000282376 Panthera tigris Species 0.000 description 1
- 240000008042 Zea mays Species 0.000 description 1
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 1
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 235000005822 corn Nutrition 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000029553 photosynthesis Effects 0.000 description 1
- 238000010672 photosynthesis Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000009663 quantitative growth Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000012358 sourcing Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000008961 swelling Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000017260 vegetative to reproductive phase transition of meristem Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to application of a remote sensing technology, in particular to a crop yield per unit and growth evaluation method based on time sequence remote sensing data, wherein the crop yield per unit evaluation method comprises the following steps: obtaining MODIS remote sensing data; carrying out normalized vegetation index extraction on the MODIS remote sensing data to construct an NDVI time sequence data set; producing a cumulative vegetation index change rate time sequence product set; acquiring a crop per unit yield statistical data set; constructing a yield per unit prediction model corresponding to different time nodes in a crop growth cycle; and (5) estimating yield of the target crops. According to the crop yield per unit and growth vigor evaluation method based on the time sequence remote sensing data, the accumulated vegetation index change rate is used as the characteristic parameter of crop yield evaluation and growth vigor monitoring evaluation, the effectiveness and the reasonability of the crop yield evaluation and the growth vigor monitoring evaluation are improved, and the method is easier to realize.
Description
Technical Field
The application relates to application of a remote sensing technology, in particular to a crop yield per unit and growth evaluation method based on time sequence remote sensing data.
Background
Under the condition that the global natural environment and the international situation are complex and changeable, the growth condition of crops is timely and accurately evaluated, and the prediction of the crop yield is crucial to national food safety management. The traditional crop yield estimation and growth evaluation adopts a manual regional investigation method, and has the advantages of low speed, large workload and high cost.
In recent years, with the development of remote sensing technology, the application of remote sensing technology in crop yield assessment and growth assessment is in a new development stage. The current crop growth remote sensing technology can be generally divided into two categories, namely a qualitative growth monitoring method and a quantitative growth monitoring method. The qualitative growth monitoring method is to use a vegetation spectral response sensitive waveband to construct remote sensing indexes capable of reflecting the growth conditions of crops, such as vegetation indexes and leaf area indexes, for monitoring.
However, in the crop yield assessment and growth vigor assessment method based on the vegetation index, due to the lack of ground observation data, the yield process prediction is not stable enough, the growth vigor evaluation standard is biased to subjectivity, and the like.
Disclosure of Invention
In order to solve at least one technical problem in the prior art, the application provides a crop yield per unit and growth evaluation method based on time sequence remote sensing data.
In a first aspect, the application discloses a crop yield evaluation method based on time series remote sensing data, which comprises the following steps:
Step 101, acquiring time sequence remote sensing data (such as MODIS remote sensing data) in a plurality of year ranges in a research area;
102, performing normalized vegetation index extraction on the time-series remote sensing data to construct an NDVI time-series data set;
step 103, extracting vegetation index characteristic values and corresponding time nodes: extracting a characteristic curve of the crop growth process by combining time sequence remote sensing data of the past year, identifying the positions of the seeding time, the lowest valley value of the mature period and the highest peak value of the growing period, and respectively recording the corresponding time periods as
And extracting the corresponding NDVImin_s、NDVImin_m、 NDVImax_gA value; and respectively obtaining the pre-growth median NDVI by an arithmetic mean method based on the extracted lowest valley value and the highest peak valuemid_gfThe corresponding time period is recorded as
And late growth median NDVImid_gbThe corresponding time period is recorded as
104, calculating the index change rate of the accumulated vegetation at different time points
firstly, calculating the vegetation index change rate of each time node; then sequentially calculate
To
To
The cumulative vegetation index change rate of any time node in between;
wherein, the vegetation index change rate and the cumulative vegetation index change rate are obtained by the following formulas:
in the formula, CRVIiIs the vegetation index change rate, NDVI, of the period iiAnd NDVIi-1Vegetation indexes of an i time period and an i-1 time period respectively, wherein i represents a time node corresponding to a time sequence product, and i-1 is a time node immediately before i; ACRVIiCumulative vegetation index change rate, t, for the end i time node0The minimum value occurrence time of the seeding period is T, and T is a specific time node;
105, acquiring the crop unit yield statistical data corresponding to the plurality of years to form a crop unit yield statistical data set;
106, performing regression analysis on the cumulative vegetation index change rate time sequence product set and the corresponding crop yield per unit statistical data set to obtain a crop yield per unit prediction model for the research area on different specific time nodes;
step 2, unit yield evaluation
Acquiring remote sensing data of the year to be evaluated, extracting vegetation index characteristic values and corresponding estimated production time nodes according to the method in the step 103, calculating vegetation index change rates and accumulated vegetation index change rates of different time nodes of the year to be evaluated according to the method in the step 104, substituting the vegetation index change rates and the accumulated vegetation index change rates into the crop per unit yield prediction model of the corresponding time nodes obtained in the step 106, and outputting a target crop per unit yield evaluation result through calculation.
According to at least one embodiment of the application, if no complete crop growth season related remote sensing data exists during estimation, effective mean characteristic images of corresponding time nodes of years of history are used as quasi-synchronization data of unfinished years, and data are updated iteratively along with growth development of crops.
It should be noted that according to the idea of the present invention, the adopted remote sensing data source can be other satellite remote sensing data besides the MODIS; in addition, in addition to the normalized vegetation index (NDVI), other characteristic parameter data reflecting the growth conditions of crops, such as Leaf Area Index (LAI), Enhanced Vegetation Index (EVI), may be employed.
According to at least one embodiment of the present application, before performing the normalized vegetation index extraction on the MODIS remote sensing data in the step 102, the method further includes:
and preprocessing the MODIS remote sensing data, wherein the preprocessing comprises splicing processing, cutting processing and quality control processing.
According to at least one embodiment of the present application, before performing the normalized vegetation index extraction on the preprocessed MODIS remote sensing data in the step 102, the method further includes:
identifying target crops in a monitoring area by using medium and high resolution remote sensing data to obtain target crop distribution data;
and carrying out mask processing on the preprocessed MODIS remote sensing data by utilizing the target crop distribution data.
According to at least one embodiment of the present application, in the step 102, after performing normalized vegetation index extraction on the masked MODIS remote sensing data, the method further includes:
filtering is performed, and Savitzky-Golay filtering is preferably used in the embodiment of the present invention.
According to at least one embodiment of the present application, the data time step is 8d, and the plurality of years refers to more than 8 years.
According to at least one embodiment of the present application, when the target crop estimation is performed in the step 2, the estimated production time node of the target crop is selected from
Wherein m is a natural number between 1 and 9.
In a second aspect, the application also discloses a crop growth evaluation method based on time series remote sensing data, which comprises the following steps:
301, obtaining a yield evaluation result of a target crop by using the crop yield evaluation method based on time series remote sensing data according to any one of the first aspect;
step 302, according to the obtained unit yield evaluation result, performing unit yield grade evaluation by combining with a unit yield grade division standard, wherein the unit yield grades comprise high yield, medium level and low yield;
and 303, estimating the growth condition of the target crops in the area to be estimated after measuring and calculating the grade standard of the growth vegetation index characteristic parameters of each specific time node according to the single yield grade division standard and the yield prediction model. Wherein, the high yield, the medium level and the low yield of the single yield grade respectively correspond to the good, medium and poor growth grades.
In accordance with at least one embodiment of the present application, in the step 302, the parity is determined as a mean value μ and a standard deviation σ of the multiple years of parity, and the (μ -0.5 σ) ≦ estimated yield ≦ (μ +0.5 σ) is graded to a medium level, the estimated yield greater than (μ +0.5 σ) is graded to a high yield, and the estimated yield less than (μ -0.5 σ) is graded to a low yield.
The innovation and the beneficial technical effects of the application are as follows:
the invention provides a new characteristic parameter for crop yield estimation and growth monitoring and evaluation: cumulative vegetation index rate of change. Application tests show that the parameters have more advantages in the aspects of crop yield estimation and growth evaluation than the accumulated NDVI and the accumulated vegetation change rate; in addition, a relative objective and convenient-to-apply long-term remote sensing monitoring evaluation mode based on yield is established.
The specific innovation points are as follows:
1) according to the crop yield per unit and growth vigor evaluation method based on the time sequence remote sensing data, the accumulated vegetation index change rate is used as the characteristic parameter for crop yield evaluation and growth vigor monitoring evaluation, so that the method is beneficial to reflecting the biomass accumulated change condition of crops in a specific growth period on one hand, can reduce the influence of crop time-space heterogeneity on the other hand, and enhances the comparability of different time-space ranges;
2) the application provides a method for extracting the index change rate characteristic parameters of the accumulated vegetation, provides a reasonable accumulation time interval range and defines a proper yield evaluation time interval (for example, the reasonable accumulation time interval range is found in the exemplary embodiment of the invention
To sequential division according to data time step
To
Any time node within; the suitable yield evaluation time interval is an interval from the time node +8d corresponding to the median value in the early growth stage to the time node +72d corresponding to the median value in the early growth stage), so that the effectiveness of the characteristic parameters of the cumulative vegetation index change rate for crop yield evaluation and growth monitoring evaluation is improved;
3) the growth quantitative evaluation method based on the predicted yield is provided, and the growth monitoring evaluation based on the mode is more objective and reasonable and is easy to realize.
Drawings
FIG. 1 is a general flow chart of a crop yield evaluation method and a growth evaluation method based on time series remote sensing data according to the present application;
FIG. 2 is a schematic diagram of a vegetation index profile for a particular crop (soybean) during growth;
FIG. 3 is a schematic spatial location of a particular study area (Toledo, west, Braziliana) of the present application;
FIG. 4 is a scatter plot of cumulative vegetation index change rate versus soybean yield at a particular time node in a particular study of the present application;
FIG. 5 is a graph showing the correlation between the index change rate of accumulated vegetation over multiple periods of time and yield per unit in a particular study of the present application;
FIG. 6 is a graphical representation of the correlation of cumulative NDVI over multiple time periods to yield per unit in a particular study of the present application;
FIG. 7 is a graph showing the correlation between the rate of change of accumulated vegetation over multiple periods of time and yield per unit area in a particular study of the present application;
FIG. 8 is a comparison graph of relative errors of index features of vegetation accumulated over multiple time periods in a particular study of the present application;
FIG. 9 is a diagram of the monitoring of the flowering and pod bearing period and the pod bearing period of soybeans in a specific year in a specific study experiment of the present application.
Detailed Description
In order to make the implementation objects, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below with reference to the drawings in the embodiments of the present application.
1. Research method
1.1 remote sensing data processing
1. Preprocessing MODIS remote sensing data based on GIS software or a related platform, wherein the preprocessing comprises splicing, cutting, quality control and the like;
2. identifying crops in a monitored area by using medium-high resolution remote sensing data, combining phenological information, interpretation mark data and the like to obtain crop distribution data, wherein soybean crops are taken as an example in the invention;
3. carrying out mask processing on the MODIS time sequence data by utilizing the crop distribution data to obtain a target crop area time sequence data set;
4. producing a normalized vegetation index (NDVI) product, and carrying out Savitzky-Golay filtering treatment to construct an NDVI time sequence data set;
NDVI is obtained from red light and near infrared bands, and because the two bands are the most important bands reflecting photosynthesis and respiration in plant spectra, the index can reflect the growth difference characteristics of vegetation, and the index is simple to calculate and easy to realize, and is widely used for monitoring the growth vigor of crops. The form is as follows:
in the formula, RnirIs the reflectivity of the near infrared band, RredThe reflectivity is in the red light wave band.
1.2 cumulative vegetation index Change Rate (ACRVI)
The vegetation growth process characterized by the change index generally comprises a positive growth stage and a negative growth stage (namely a descending stage), the growth trend of the vegetation can be characterized by utilizing the difference value of the indexes of the vegetation before and after the vegetation, and in order to remarkably characterize the vegetation change degree and the vegetation change trend at a certain time point, the research provides the vegetation index change rate, and the form is shown in a formula (2); in order to represent the variation degree of the vegetation growth process, the accumulated vegetation index variation rate is provided, and the form is shown in formula (3):
in the formula, CRVIiIs the vegetation index change rate, NDVI, of the period iiAnd NDVIi-1Vegetation indexes of an i time period and an i-1 time period respectively, wherein i represents a time node corresponding to a time sequence product, and i-1 is a time node immediately before i;
ACRVIicumulative vegetation index change rate, t, for the end i time node0As the initial time, the lowest value of the emergence time of the sowing period is taken as t in the study0And T is a specific time node.
1.3 cumulative vegetation index change rate eigenvalue extraction
Extracting 5 vegetation index characteristic values including minimum valley value (marked as NDVI) of sowing period by combining with crop growth process curvemin_s) The highest peak of growth period (designated as NDVI)max_g) The lowest-highest median value in the pre-growth period (abbreviated as "median value in the pre-growth period", and designated as NDVI)mid_gf) The highest-lowest median value in the late stage of growth (called the median value in the late stage of growth, NDVI)mid_gb) And the lowest trough in maturity (noted NDVI)min_m)。
Firstly, determining a monitoring start-stop time period (recorded in months) according to the soybean planting habits in a research area;
secondly, performing position identification on the seeding time, the lowest valley value of the mature period and the highest peak value of the growing period by adopting a secondary difference method, and respectively recording the corresponding time periods as
Extracting corresponding NDVI values, and respectively obtaining a median value in the early stage of growth (the corresponding time period is recorded as the corresponding time period) by using an arithmetic mean method based on the extracted lowest valley value and the highest peak value
) And late growth median (its corresponding interval is noted as
)；
Thirdly, calculate
To
CRVI value of (a);
finally, to
Counting the starting point for the cumulative vegetation index change rate to
For counting the end point, calculating
And obtaining the cumulative vegetation index change rate sequences of different years by different time points by dividing the cumulative vegetation index change rate ACRVI of different time nodes according to the data time step in the statistical interval, wherein the data time step is 8 d.
It should be noted that typically crop yield predictions and growth assessments are made during the crop growing season, i.e., there may not be complete crop growing season related remote sensing data, and therefore, this study is referenced to Hoolst R V et al[1]The method obtains effective mean characteristic images of different time points of years, takes the effective mean characteristic images as quasi-synchronous data of the unfinished year, and carries out iterative updating of the data along with the growth development of crops.
1.4 other feature parameter extraction
In order to objectively evaluate the yield estimation and growth estimation effects based on the cumulative vegetation index change rate, two common characteristic parameters, namely the cumulative NDVI and the cumulative vegetation index change rate, are obtained through synchronous calculation in the research. The calculation formula is as follows:
in the formula, ANDVIiCumulative vegetation index, t, for the node by i time0As the initial time, the lowest value of the emergence time of the sowing period is taken as t in the study0；SRVIiThe vegetation index change rate of the time node in the period i is obtained, and t is the time interval between the time node in the period i and the time node in the period i-1; ASRVIiThe cumulative vegetation index change rate of the i time node is used as a cutoff.
1.5 precision analysis method
In order to evaluate the accuracy of the model, the model is analyzed and evaluated by using a common model test index. The coincidence degree between the model simulation value and the measured value is analyzed by calculating the relative error and the determination coefficient between the model simulation value and the measured value. The calculation formula is as follows:
in the formula, ReAs a relative error, yiAnd y is a predicted value and an actual value, respectively; r2To determine the coefficients, n is the number of samples. R2The value range is (0, 1), and a value closer to 1 indicates a higher degree of agreement between the analog value and the measured value, whereas a value closer to 0 indicates a lower degree of agreement.
An exemplary application:
1) area of investigation
The research area is Toledo city located in the west of Balana, Brazil, the longitude range of the research area is-54 degrees 3 '-53 degrees 32', the latitude range is-24 degrees 27 '-24 degrees 56', the altitude is 254-690 m, the average altitude is 503m, and the area of the county is 1196.2km2(FIG. 3). The temperature of toledo is generally between 9.4 ℃ and 30 ℃. Fertile soil and relatively flat terrain make this market an agricultural development. It is one of the important food production bases in this state, the agricultural GDP is ranked first in Balana and south regions, and the agricultural added value is listed third in the national rank[2]. The main crop types are soybean, corn, tobacco leaf, coffee, etc. Wherein the growing season of the soybean is 9 months to 3 months in the next year.
2) Data sourcing and processing
2.1) statistical investigation data
Statistics data of annual single yield of soybeans in Toledo city and data of seedling condition progress and growth survey are from the department of economy in rural areas (DERAL) in Balana, Brazil. Wherein the used statistical data age per unit yield is 2010/11-2020/21, and the seedling condition progress and growth survey data age is 2019/20-2020/21.
2.2) remote sensing and basic geographic data
In this study, a MOD09A1 (surface reflectance) 8d synthetic product using a mode Resolution Imaging spectrometer (MODIS) was synthesized, the spatial Resolution was 500m, and the data time was 9 months in 2010 to 3 months in 2021. And carrying out preprocessing such as splicing, cutting, quality control and the like on the composite material. The forecasting result has the highest accuracy rate after the crop mask is used[3]Therefore, the soybean planting region extraction is performed on the research region by using the Landsat8 data based on a machine learning method, and the soybean planting region extraction is used as mask data to complete the mask processing. Then normalized vegetation index (NDVI) extraction is carried out, and the vegetation index is subjected to Savitzky-Golay[4]And after filtering, generating the NDVI time sequence product of the soybean production area.
The temporal expression of the NDVI time series products in this study was based on the expiration date of the 8d synthesis.
2.3) extracting vegetation index characteristic value and corresponding specific time node
Extracting a crop growth process characteristic curve by combining time sequence remote sensing data of 9 months to 2021 months in 2010, carrying out position identification on the seeding time, the lowest valley value of the mature period and the highest peak point of the growing period of each year, and respectively recording the corresponding time periods as
And extracting the corresponding NDVImin_s、NDVImin_m、NDVImax_gNDVI values; and respectively obtaining the pre-growth median NDVI by an arithmetic mean method based on the extracted lowest valley value and the highest peak valuemid_gfThe corresponding time period is recorded as
And late growth median NDVImid_gbThe corresponding time period is recorded as
2.4) calculating the cumulative vegetation index change rate of nodes at different times
The cumulative vegetation index change rate is obtained by the following formula:
in the formula, CRVIiIs the vegetation index change rate, NDVI, of the period iiAnd NDVIi-1Vegetation indexes of an i time period and an i-1 time period respectively, wherein i represents a time node corresponding to a time sequence product, and i-1 is a time node immediately before i; ACRVIiCumulative vegetation index change rate, t, for the end i time node0T is a specific time node for the lowest value occurrence time of the seeding time.
And 106, performing regression analysis on the cumulative vegetation index change rate time sequence product set and the corresponding crop yield per unit statistical data set to obtain a crop yield per unit prediction model for the research area at different specific time points.
3) Results and analysis of the study
3.1) analysis of relation between index characteristic parameters of accumulated vegetation and yield
3.11)
And
analysis of relation between accumulated vegetation index characteristic parameter and yield per unit
For 2010/11-2018/19 years
Cumulative vegetation index change rate and soybean yield per unitThe line regression analysis, as shown in fig. 4(a), showed that the cumulative vegetation index change rate exhibited a positive linear relationship with soybean yield per unit.
Calculating a correlation coefficient of 0.88 using Pearson correlation analysis, determining a coefficient of 0.77, and passing a significance test with a significance level of 0.01; the relative errors obtained by predicting the unit yields of 2019/20 and 2020/21 in two years by using a linear regression model are-2.51% and 2.14% respectively, and the average relative error is-0.18%.
The correlation coefficient of the cumulative NDVI with soybean per unit area was 0.79, the determination coefficient was 0.62, and the significance test with significance level of 0.05 was passed. Using a linear regression model, the predicted yield per unit of 2019/20 and 2020/21 years was found to have relative errors of 13.97% and 27.20%, respectively, and an average relative error of 20.59%.
The correlation coefficient of the cumulative vegetation index change rate and the soybean yield per unit is 0.89, the determination coefficient is 0.8, and the significance test with the significance level of 0.01 is passed. Using a linear regression model, the predicted yield per unit of 2019/20 and 2020/21 years was obtained with relative errors of 10.74% and 0.78%, respectively, and an average relative error of-5.76%.
The comparison shows that the yield prediction effect is optimal based on the cumulative vegetation index change rate in the period, the cumulative vegetation index change rate prediction effect is inferior, and the cumulative NDVI prediction effect is poor.
For 2010/11-2018/19 years
The cumulative vegetation index change rate and the soybean yield per unit are subjected to regression analysis, and as shown in fig. 4(b), the cumulative vegetation index change rate and the soybean yield per unit show a negative linear relationship.
Correlation coefficient was calculated to be 0.54 using Pearson correlation analysis, with a decision coefficient of 0.288, failing the test with significance level of 0.05. The unit yields of 2019/20 and 2020/21 are predicted by using a linear regression model, the relative errors are respectively-21.48% and-4.48%, and the average relative error is-12.98%.
The correlation coefficient of the cumulative NDVI with soybean per unit area was 0.79, the decision coefficient was 0.63, and the significance test with a significance level of 0.05 was not passed. Using a linear regression model, the predicted yield per unit of 2019/20 and 2020/21 years was obtained with relative errors of-8.82% and 4.65%, respectively, and an average relative error of-2.09%.
The correlation coefficient of the cumulative vegetation index change rate and the soybean yield per unit is 0.77, the determination coefficient is 0.59, and the significance test with the significance level of 0.05 is passed. Using a linear regression model, the predicted yield per unit of 2019/20 and 2020/21 years was obtained with relative errors of-18.62% and-0.91%, respectively, and an average relative error of-9.77%.
In the stage, only the change rate of the accumulated vegetation index passes the significance test, but the average relative error is higher, and although the average relative error of the accumulated NDVI is lower and better than the fitting effect of the other two characteristic parameters, the average relative error does not pass the significance test, so that the yield prediction of the three characteristic parameters is not suitable in the stage.
3.12) analysis of relation between multi-period accumulative vegetation change characteristic parameters and yield per unit
Corresponding time node with the median value in the early growth stage
The start time is added with 8d in sequence until 80d is added to calculate the end time of the cumulative vegetation change characteristic, 11 monitoring time patterns are counted, and correlation analysis is carried out on the three cumulative vegetation index change characteristic of 2010/11-2018/19 for 9 years of single soybean production and the 11 monitoring time patterns, and the result is shown in fig. 5-7 (wherein, the is extremely significant correlation and the is significant correlation).
The cumulative vegetation index change rate and the cumulative vegetation index change rate are similar to the distribution on the time sequence scale of the correlation coefficient per unit yield, are in a parabolic shape, are low in the early stage and the later stage, the correlation is not significant, the significance level is significant or extremely significant along with the development of the growth stage, and the correlation reaches a peak value when the median corresponds to a time node +40d in the early stage of growth, and is respectively 0.96 and 0.97. The accumulated NDVI shows a linear increasing trend in general except that a low value appears when the median corresponds to a time node +24d in the early growth stage, namely, the correlation coefficient gradually rises along with the development of the growth stage, and reaches a high value of 0.89 in the late growth stage, and the NDVI shows extremely significant correlation.
The median value in the early growth stage is mainly the soybean seedling stage-branch stage, when the soybean seedling stage is just started, the three characteristic parameters of the cumulative vegetation index change rate, the cumulative NDVI and the cumulative vegetation index change rate and the yield correlation coefficient are all lower, and the correlation relationship does not reach the significant level. As soybeans grow, the cumulative vegetation index change rate and the cumulative vegetation index change rate have a significant or extremely significant correlation with yield per unit, the correlation coefficient of the cumulative vegetation index change rate and the cumulative vegetation index change rate are higher than that of the cumulative vegetation index change rate and can be considered to be in the period from the time node +8d corresponding to the early-stage median value to the time node +72d corresponding to the early-stage median value, namely TNDVImid_gf+8d～ TNDVImid_gfDuring the period of +72d, the yield per unit of soybean is well reflected, can be used for estimating the yield of the soybean and is a preferred index in three characteristic parameters. In addition, the accumulated NDVI and soybean yield correlation coefficient is better than the other two characteristic parameters in the time node +72d corresponding to the median in the early growth stage and later, and can be used as a yield estimation parameter in the whole growth stage.
Further, for 2010/11-2018/19 years
The accumulated vegetation index characteristic parameters and the soybean yield are subjected to regression analysis, linear fitting models in different periods are established, the yields in two years of 2019/20 and 2020/21 are predicted based on the models, and the relative error results of the yield prediction based on the three accumulated vegetation index characteristic parameters are obtained, as shown in fig. 8:
the prediction results based on the accumulated NDVI are overall high and have high relative errors, wherein 2019/20 is used for predicting the relative error between-13.53% and 36.63%, and 2020/21 is used for predicting the relative error between-0.07% and 43.71%.
The prediction result based on the change rate of the cumulative vegetation index is low overall, and the relative error in the middle and later stages of soybean growth is large in difference of two years, wherein the 2019/20 prediction relative error is-17.76% -4.33%, and the 2020/21 prediction relative error is-7.35% -5.3%.
The prediction result based on the index change rate of the accumulated vegetation is higher in the early stage and lower in the later stage, wherein the 2019/20 prediction relative error is-13.52% -14.65%, and the 2020/21 prediction relative error is-7.43% -6.98%.
Relative errors of 2019/20 and 2020/21 in two years are averaged to obtain relative error mean value time sequence curves of three characteristic parameters, wherein the relative error mean value based on the accumulated NDVI is between-6.8% and 40.2%, the relative error mean value based on the change rate of the accumulated vegetation index is between-6.8% and 6.6%, and the relative error mean value based on the change rate of the accumulated vegetation index is between-11.0% and-0.6%.
Further, the relative error of the yield prediction of 2019/20 and 2020/21 based on three accumulated vegetation index characteristic parameters in two years is averaged in a time period of
A period of time is
The results are shown in Table 1.
TABLE 1 comparison of average relative errors of multiple-interval accumulated vegetation index characteristic parameters in different intervals
It can be seen that: the predicted relative error based on the accumulated NDVI was overall highest, with an average relative error of 23.28% over multiple periods in 2019/20 years and 32.74% in 2020/21 years.
The smallest average relative error is the index change rate of the accumulated vegetation, wherein the average relative error of 2019/20 years in multiple periods is 0.48%, and the average relative error of 2020/21 years is 0.71%. The predicted effect based on the rate of change of the cumulative vegetation index was second, with 2019/20 being-6.96% and 2020/21 being-1.6%.
Due to the fact that
Most of the soybean blossoming and pod bearing-grain swelling period is a key stage of soybean yield development, so that the predicted average relative error of the stage is compared and analyzed.
The predicted relative error based on the cumulative NDVI was overall highest with an average relative error of 33.22% in 2019/20 years and 40.65% in 2020/21 years. The smallest average relative error is the cumulative vegetation index change rate, wherein the average relative error at 2019/20 years is 0.11%, and the average relative error at 2020/21 years is 3.69%. The predicted effect based on the rate of change of the cumulative vegetation index was second, with 2019/20 being-6.42% and 2020/21 being 1.17%.
3.2) typical annual growth analysis based on the characteristic parameters of the index variations of the cumulative vegetation
Based on the analysis result of the relation between the index change characteristic parameters of the accumulated vegetation and the yield per unit, the index change characteristic parameters of the accumulated vegetation can reflect the future yield level and trend of crops within a specific prediction precision range, so that the related parameters can be used for monitoring and evaluating the growth in different periods.
And selecting the lowest valley-highest peak time period of typical years of 2010/11-2020/21, including 2016/17 (the highest yield year), 2019/20 (the last year) and 2020/21 (the current year) for application and analysis.
The growth grade of each specific time node needs to be calculated based on a single-yield grade and a single-yield prediction model, wherein the single-yield grade is determined by a multi-year single-yield mean value mu and a standard deviation sigma, and in the research, the yield is graded in a [ mu-0.5 sigma, mu +0.5 sigma ] interval to be a medium level (the corresponding growth grade is medium), graded to be better than (mu +0.5 sigma) (the corresponding growth grade is good), and graded to be worse than (mu-0.5 sigma) (the corresponding growth grade is poor).
Based on the standard, the growth evaluation result of the three year highest peak periods based on the index change characteristic parameter of the accumulated vegetation is obtained, as shown in fig. 9.
Comparative analysis and relative error analysis were performed using DERAL2019/20 and 2020/21 two year contemporaneous survey statistics, as shown in Table 2 below.
TABLE 2 growth rating and accuracy statistics
According to statistical data, 3846kg/ha was obtained in 2019/20, and 3350kg/ha was obtained in 2020/21. According to the year, the growth monitoring chart (figure 9) based on the three accumulated vegetation index change characteristic parameters can show that the growth of 2020/21 years is shorter than that of the first two years in general, and the good proportion of the annual seedling condition is lower than 2019/20 according to table 2. The reason is that the soybean planting and seedling emergence period in the year has severe drought, so that the sowing is late and the seedling condition is weak in partial areas, the biomass accumulation is influenced, and the growth performance is biased.
In order to evaluate the growth evaluation effect based on the three accumulated vegetation index change characteristic parameters, the relative error measurement and calculation are carried out by combining statistical survey data.
As can be seen from table 2, the relative errors at 2019/20 years based on the cumulative vegetation index rate of change are 0.5%, 1.3%, and-4.4%, respectively; the relative errors at 2020/21 years were-1.9%, -0.2%, and 23.1%, respectively. In two years, the average relative errors of the good and medium grades are-0.7% and 0.5%, respectively, and the difference grade is lower, 9.4%, and overall, the relative error is lower than 10%, especially the relative errors of the good and medium grades are very low, between-1.9% and 1.3%.
Relative errors at 2019/20 years based on cumulative NDVI were 7.6%, 4.3%, and-53.4%, respectively; the relative errors at 2020/21 years were 20.9%, -47.8%, and-52.3%, respectively. The relative error between the different levels is more variable in two years, particularly between the medium and poor levels, due to the tendency of the cumulative NDVI-based estimates to be higher in the perennial estimates and lower in the high-yield yearly estimates, resulting in greater variation in the growth assessments.
Relative errors at 2019/20 years based on the rate of change of the cumulative vegetation index were-3.4%, 29.1%, and 3.0%, respectively; the relative errors at 2020/21 years were-3.6%, 9.8%, and 2.6%, respectively. In two years, the average relative errors for the good and poor grades were-3.5% and 2.8%, respectively, and for the medium grade, 19.5% was low. The evaluation effect is between the first two characteristic parameters.
In summary, the cumulative vegetation index change rate is the most feasible and effective growth assessment parameter of the three characteristic parameters.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Reference documents:
[1]Hoolst R V,Eerens H,Haesen D,et al.FAO's AVHRR-based Agricultural Stress Index System(ASIS)for global drought monitoring[J].International Journal of Remote Sensing,2016,37(1-2):418-439.
[2]Manfrin,J.,Ceraso,A.F.F.,&
Jr,A.C.Evaluation of Odours Perception by Residents of the Municipality of Toledo/ParanáState/Brazil[J].Journal of Scientific Research and Reports,2018,21(2),1-9.https://doi.org/10.9734/JSRR/2018/44890.
[3] extraction and yield estimation of rice development period based on time series vegetation index [ D ]. Zhejiang university, 2017.
[4] The reconstruction [ J ] of the remote sensing report, 2010,14(04):725-741 by Savitzky-Golay filtering algorithm of the edge gold tiger, Liainong, Song Meng, et al.MODIS vegetation index time series.
Claims (9)
1. A crop yield evaluation method based on time sequence remote sensing data comprises the following steps:
step 1, constructing a crop yield per unit prediction model
Step 101, acquiring time sequence remote sensing data in a plurality of year ranges in a research area;
102, extracting Normalized Difference Vegetation Index (NDVI) from the time-series remote sensing data to construct an NDVI time-series data set;
step 103, extracting vegetation index characteristic values and corresponding time nodes: extracting a characteristic curve of the crop growth process by combining time sequence remote sensing data of the past year, identifying the positions of the seeding time, the lowest valley value of the mature period and the highest peak value of the growing period, and respectively recording the corresponding time periods as
And extracting the corresponding NDVImin_s、NDVImin_m、NDVImax_gA value; based on the extracted lowest valley value and highest peak value, respectively obtaining the median NDVI in the early stage of growth by using an arithmetic mean methodmid_gfThe corresponding time period is recorded as
And late growth median NDVImid_gbThe corresponding time period is recorded as
104, calculating the cumulative vegetation index change rate of nodes at different time;
firstly calculating the vegetation index change of each time nodeRate; then sequentially calculate
To
To
The cumulative vegetation index change rate of any time node in between;
wherein, the vegetation index change rate and the cumulative vegetation index change rate are obtained by the following formulas:
in the formula, CRVIiIs the vegetation index change rate, NDVI, of the period iiAnd NDVIi-1Vegetation indexes of an i time period and an i-1 time period respectively, wherein i represents a time node corresponding to a time sequence product, and i-1 is a time node immediately before i; ACRVIiCumulative vegetation index change rate, t, for the end i time node0The minimum value occurrence time of the seeding period is T, and T is a specific time node;
105, acquiring the crop unit yield statistical data corresponding to the plurality of years to form a crop unit yield statistical data set;
106, performing regression analysis on the cumulative vegetation index change rate time sequence product set and the corresponding crop yield per unit statistical data set to obtain a crop yield per unit prediction model for the research area on different time nodes;
step 2, unit yield evaluation
Acquiring remote sensing data of the year to be evaluated, extracting vegetation index characteristic values and corresponding estimated production time nodes according to the method in the step 103, calculating vegetation index change rates and accumulated vegetation index change rates of different time nodes of the year to be evaluated according to the method in the step 104, substituting the vegetation index change rates and the accumulated vegetation index change rates into the crop per unit yield prediction model of the corresponding time nodes obtained in the step 106, and outputting a target crop per unit yield evaluation result through calculation.
2. The method of claim 1, wherein if complete crop growth season-related remote sensing data is lacking in the single-yield evaluation, the effective mean feature image of the time node corresponding to the years of the history is used as quasi-contemporaneous data of the unfinished year, and the data is iteratively updated as the crop growth period progresses.
3. The method of claim 1, wherein before performing the normalized vegetation index extraction on the MODIS remote sensing data in the step 102, the method further comprises:
and preprocessing the MODIS remote sensing data, wherein the preprocessing comprises splicing processing, cutting processing and quality control processing.
4. The method of claim 1, wherein before the step 102 of performing the normalized vegetation index extraction on the preprocessed MODIS remote sensing data, the method further comprises:
identifying target crops in a monitoring area by using medium and high resolution remote sensing data to obtain target crop distribution data;
and carrying out mask processing on the preprocessed MODIS remote sensing data by utilizing the target crop distribution data.
5. The method of claim 4, wherein after the step 102 of performing the normalized vegetation index extraction on the masked MODIS remote sensing data, the method further comprises: and (5) filtering.
6. The method of claim 1, wherein the data time step is 8d, and the plurality of years refers to more than 8 years.
7. The method according to claim 6, wherein the estimated production time node of the target crop is selected from the group consisting of when the target crop is estimated in step 2
Wherein m is a natural number between 1 and 9.
8. A crop growth evaluation method based on time sequence remote sensing data comprises the following steps:
301, obtaining a yield evaluation result of a target crop in an area to be evaluated by adopting the time-series remote sensing data-based crop yield evaluation method according to any one of claims 1 to 7;
step 302, according to the obtained unit yield evaluation result, performing unit yield grade evaluation by combining a unit yield grade division standard, wherein the unit yield grade is divided into high yield, medium level or low yield;
and 303, after measuring and calculating the grade standard of the growth vegetation index characteristic parameter of each specific time node according to the single-yield grade division standard and the yield prediction model, evaluating the growth condition of the target crop in the area to be evaluated, wherein the high yield, the medium level and the low yield of the single-yield grade respectively correspond to the good, medium and poor growth grades.
9. The method of assessing the growth of a crop as claimed in claim 8, wherein in said step 302, the yield rating is determined as the mean μ per year and the standard deviation σ, and wherein (μ -0.5 σ) is evaluated as a medium level with yield ≦ μ +0.5 σ, and wherein (μ +0.5 σ) is evaluated as a high yield and (μ -0.5 σ) is evaluated as a low yield.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111200212.0A CN114118679B (en) | 2021-10-14 | 2021-10-14 | Crop yield per unit and growth evaluation method based on time sequence remote sensing data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111200212.0A CN114118679B (en) | 2021-10-14 | 2021-10-14 | Crop yield per unit and growth evaluation method based on time sequence remote sensing data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114118679A true CN114118679A (en) | 2022-03-01 |
| CN114118679B CN114118679B (en) | 2022-09-16 |
Family
ID=80375655
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111200212.0A Active CN114118679B (en) | 2021-10-14 | 2021-10-14 | Crop yield per unit and growth evaluation method based on time sequence remote sensing data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114118679B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115511224A (en) * | 2022-11-11 | 2022-12-23 | 中国农业科学院农业资源与农业区划研究所 | Intelligent monitoring method and device for growing vigor of crops integrated with heaven and earth and electronic equipment |
| CN116976516A (en) * | 2023-08-01 | 2023-10-31 | 中国科学院空天信息创新研究院 | Early prediction method for single crop yield in arid region |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106772429A (en) * | 2016-12-24 | 2017-05-31 | 福州大学 | Increase and decrease the corn autodraft method of Ratio index based on peak of growing season NMDI |
| CN108362267A (en) * | 2018-01-09 | 2018-08-03 | 浙江大学 | The lower yield of rape of wet stain evil stress based on satellite data loses quantitative evaluation by remote sensing method |
| CN110309985A (en) * | 2019-07-10 | 2019-10-08 | 北京师范大学 | A kind of crop yield prediction technique and system |
| WO2020132092A1 (en) * | 2018-12-19 | 2020-06-25 | The Board Of Trustees Of The University Of Illinois | Apparatus and method for crop yield prediction |
| WO2021007352A1 (en) * | 2019-07-08 | 2021-01-14 | Indigo Ag, Inc. | Crop yield forecasting models |
-
2021
- 2021-10-14 CN CN202111200212.0A patent/CN114118679B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106772429A (en) * | 2016-12-24 | 2017-05-31 | 福州大学 | Increase and decrease the corn autodraft method of Ratio index based on peak of growing season NMDI |
| CN108362267A (en) * | 2018-01-09 | 2018-08-03 | 浙江大学 | The lower yield of rape of wet stain evil stress based on satellite data loses quantitative evaluation by remote sensing method |
| WO2020132092A1 (en) * | 2018-12-19 | 2020-06-25 | The Board Of Trustees Of The University Of Illinois | Apparatus and method for crop yield prediction |
| WO2021007352A1 (en) * | 2019-07-08 | 2021-01-14 | Indigo Ag, Inc. | Crop yield forecasting models |
| CN110309985A (en) * | 2019-07-10 | 2019-10-08 | 北京师范大学 | A kind of crop yield prediction technique and system |
Non-Patent Citations (2)
| Title |
|---|
| ZHONGLIN JI 等: "Prediction of crop yield using phenological information extracted from remote sensing vegetation index", 《SENSORS》 * |
| 陈鹏飞 等: "基于环境减灾卫星时序归一化植被指数的冬小麦产量估测", 《农业工程学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115511224A (en) * | 2022-11-11 | 2022-12-23 | 中国农业科学院农业资源与农业区划研究所 | Intelligent monitoring method and device for growing vigor of crops integrated with heaven and earth and electronic equipment |
| CN116976516A (en) * | 2023-08-01 | 2023-10-31 | 中国科学院空天信息创新研究院 | Early prediction method for single crop yield in arid region |
| CN116976516B (en) * | 2023-08-01 | 2024-03-15 | 中国科学院空天信息创新研究院 | Early prediction method for single crop yield in arid region |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114118679B (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fischer et al. | Global agro-ecological zones (gaez v4)-model documentation | |
| CN114118679B (en) | Crop yield per unit and growth evaluation method based on time sequence remote sensing data | |
| CN105760978B (en) | One kind being based on the agricultural drought disaster grade monitoring method of temperature vegetation drought index (TVDI) | |
| CN109711102B (en) | Method for rapidly evaluating crop disaster loss | |
| CN111598019A (en) | Crop type and planting mode identification method based on multi-source remote sensing data | |
| CN109800921B (en) | Regional winter wheat yield estimation method based on remote sensing phenological assimilation and particle swarm optimization | |
| Okada et al. | Projecting climate change impacts both on rice quality and yield in Japan | |
| Zhao et al. | Normalized NDVI valley area index (NNVAI)-based framework for quantitative and timely monitoring of winter wheat frost damage on the Huang-Huai-Hai Plain, China | |
| Joshi et al. | Winter wheat yield prediction in the conterminous United States using solar-induced chlorophyll fluorescence data and XGBoost and random forest algorithm | |
| Niu et al. | A 30 m annual maize phenology dataset from 1985 to 2020 in China | |
| Rao et al. | Remote sensing: A technology for assessment of sugarcane crop acreage and yield | |
| CN111985728A (en) | Method for establishing organic sorghum yield prediction model | |
| WO2024099475A1 (en) | Soil heavy metal stress identification method based on long time series ndvi | |
| CN107437262B (en) | Crop planting area early warning method and system | |
| CN117077868A (en) | Model-based grain yield prediction system and method | |
| CN109615150B (en) | Method and system for determining rice meteorological output | |
| Bhimani et al. | Forecasting of groundnut yield using meteorological variables | |
| Sudianto et al. | Early warning for sugarcane growth using phenology-based remote sensing by region | |
| CN111626638A (en) | Construction and application of summer corn lodging meteorological grade evaluation model | |
| Ji et al. | Using NDVI time series curve change rate to estimate winter wheat yield | |
| Garde et al. | Weather based pre-harvest forecasting of wheat at Ghazipur (UP) | |
| Mech | Status of tea production in Assam: past trends and its future projections | |
| Gill et al. | Development of agrometeorological models for estimation of cotton yield | |
| CN117744898B (en) | Construction method of annual prediction model of yield of field grain crops | |
| CN111667167B (en) | Agricultural grain yield estimation method and system |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240314 Address after: 100125 Building 2, national agricultural exhibition hall, No. 16, North East Third Ring Road, Chaoyang District, Beijing Patentee after: Big data development center of the Ministry of agriculture and rural areas Country or region after: China Address before: 100125, No. 41, wheat Street, Chaoyang District, Beijing Patentee before: Ministry of Agriculture and Rural Planning and Design Institute Country or region before: China |
|
| TR01 | Transfer of patent right |