Crop Classification Based On Analysis Of Phenological Characteristics Of MODIS Timesserise

China is an agricultural country, crop planting area is one of the important factors affecting food production, rapid, timely and accurate crop distribution information, has become one of the great modernized agriculture development requirement, and the identification and classification of crops, the social economy, food security, ecological function and policy have important influence.Firstly, based on the Asymmetric Gaussians method of TIMESAT software, the MOD09Q1 datasets with 250 m resolution was used to filter and reconstruct the time-series NDVI curves. Then the seven phenological characteristics(start time of the season; end time of the season; Length of the season; seasonal amplitude; rate of increase at the beginning of the season; rate of decrease at the end of the season; small seasonal integral) were extracted. Secondly, to analyze the time-series NDVI curve characteristics of vegetables, water and construction land were masked off because their maximum NDVI values were less than 0.5. Then in order to get the optimal classification accuracy of the crop land, Hierarchical classification method were conducted as below:(1) using SVM classification to extract agricultural area based on the time-series NDVI data;(2) using SVM classification to identify three crop classes(soybean, corn, rice) with different combination of three bands(NDVI: NDVI bands; PH: phonological bands; NDWI: NDWI bands) on the basis of the first step.After the above process is concluded as follows:(1)TIMESAT software can effectively to smooth denoising of NDVI time-series, can make the NDVI curve in keeping the original basic shapes contained more effectively on the basis of revealing the phenology of cyclical change rule, highlighting the advantages of MODIS timing and the extraction of phenological parameters have great help.(2)Using SVM to NDVI time-series extract of farmland classification results of agricultural land in mapping accuracy is 97.78%, the user accuracy of 95.65%.The rest of the three types of precision is also higher, the NDVI time-series to extract the information of farmland is feasible.(3) We compare the value of Overall Accuracy and Kappa coefficient of different combination, the result was returned as below: NDVI+NDWI>NDVI+PH+NDWI>PH+NDWI>NDVI+PH>NDVI>PH. Finally, the higher dimensions won’t bring the higher accuracy necessarily, then finally that the application of NDWI can enhance the Overall accuracy of rice effectively. At the same time, it is workable to identify the crop types in large scale combining phonological information with other character datasets in this thesis.(4)According to statistical classification results,we know the area of corn is more broad than soybean and rice.