| The identification and acquisition of crop planting information is the basic work of agricultural remote sensing and agricultural monitoring.The key is to identify and separate crops from many ground objects.Remote sensing data of various resolutions today can provide a good data base for monitoring crop spatial information,especially the multi-temporal vegetation index and some other spectral characteristics,which provide a basis for remote sensing monitoring and identification of crop growth season changes.Compared with traditional crop planting area monitoring methods,remote sensing extraction methods have the advantages of wider scope,higher efficiency and lower cost.Therefore,quantitative monitoring of crops is effective and feasible.The use of remote sensing technology can achieve dynamic,efficient and accurate access to spatial information of crops,help the relevant departments of the state to macro-control and adjust the agricultural structure and formulate scientific and rational policies,and make rational selection of crops for farmers to obtain dynamic monitoring information.Risk avoidance and other aspects are of great significance.In this paper,Henan Province was used as the research area,and the MODIS double-satellite 8-day time-resolution vegetation index product was used as the original data.After smoothing the data,combined with crop phenology,multiple methods were used to extract layers,which was obtained in 2005,2008 2010,2013 and 2015,the multiple cropping index of Henan Province and the spatial distribution of main crops,and the accuracy of the study area and the typical area were verified by the extraction results,which further studied the method of extracting crop information,and hoped to provide for the extraction of crop information.Some references.The research contents and results of this thesis are as follows:(1)Extraction of the multiple cropping index of Henan Province.Using the double-star data of the 8day time series of MODIS NDVI,the smoothed data is obtained by Savizky-Golay filtering,and the NDVI time series curves of the main land,water,forest and cultivated land are extracted,and the NDVI is judged and analyzed.The rule is to use the second difference method to extract the NDVI to obtain the peak data on each pixel.After the peak is selected,the crop distribution of the crops of the year and the season of the year is obtained,and the multiple cropping index in the traditional sense is obtained.Determine the accuracy.The overall relative error of the multiple cropping index calculated by remote sensing is kept within 10%.Overall,the results of the multiple cropping index extracted by remote sensing show a high correlation at the municipal level,indicating that the method in this study is feasible.Basically reliable.(2)Analysis of the spatial distribution pattern of multiple cropping indices.In terms of distribution,Henan Province mainly has a two-year and one-year-a-year system.The two-year-old cultivated land is mainly distributed in the central and eastern plains of Henan and the Nanyang Basin.The annual cultivated land is mainly distributed in the western mountains.The area and the southern Xinyang area.The multiple cropping index of the eastern city in space is higher than the multiple cropping index of Xinyang in the west and south.The average multiple cropping index for Henan Province in the past five years was 183.6%.The overall multiple cropping index change in Henan Province in the past 10 years was from 1.919 in 2005to 183.7%in 2010.The rate of decline was faster,and the rate of decline was 4.46%.From 2010 to 2015.The multiple cropping index remained at a stable level,becoming 188.3%in 2015.(3)Extract the spatial distribution of the main crops.Using MODIS NDVI time series data to calculate the absolute dispersion,select the most suitable method for the threshold selection,obtain the suspected distribution map of different crops,and use the reflectivity data of MODIS,and use the mixed pixel decomposition to obtain the main crop pixels.Abundance,finally verified the accuracy of the study area and the training area.The results show that the accuracy of winter wheat identification is 85.63%in the accuracy verification of statistical data,the accuracy of summer maize identification is 78.07%,and the accuracy in random point verification reaches 83.02%.(4)Analysis of the spatial distribution of major crops.The results of the study indicate that winter wheat and summer maize are the main food crops in Henan.The average planting area of winter wheat and summer maize in Henan Province in 2005,2008,2010,2013 and 2015 was 4567.33 hm~2 and 2349.46 hm~2.In the interannual spatial distribution,there is a clear pattern of consistency,mainly distributed in the relatively flat areas of Henan,with less distribution in the mountains and hills.In terms of quantitative changes,the total planting area of these two crops showed a significant increase trend during the 10 years.The rising rate of winter wheat was 9.26%,and the rising rate of summer maize was 33.31%. |
PDF Full Text Request