The Extraction Of Beijing Main Cops Planting Area Based On Time Series Modis Ndvi Reconstruction

In this paper, the author explored an approach using time series Normalize Difference Vegetation Index (NDVI) derived from MODIS to extract planting area of main crop in Beijing. Considering the high temporal resolution advantage of MODIS data, MODIS is a potential remotely sensed data source to monitor crop-planting information. Firstly, MODIS data of thirteen key phenological stages from March 21 to Sep 29 were selected to classify the different crops in Beijing. Secondly, in order to remove cloud contamination, the time series of MODIS NDVI data are filtered by the Harmonic Analysis of Time Series (HANTS) algorithm. The HANTS algorithm offers greater flexibility in the choice of frequencies and the length of the time series than the FFT algorithm. Thirdly, considering the effects of topography and geographical latitude on crop growth stages, the study area was divided by GIS ancillary data. Finally, based on the NDVI temporal spectral features of different crop in Beijing, a decision tree classification algorithm is designed, and the planting areas of main crop in Beijing are successfully extracted by the algorithm.