Cropland classification from MODIS-Landsat fusion data

摘要

Crop mapping requires information of crop phenology regarding the high spatiotemporal resolution satellite data. This study aims to develop an approach integrating the Moderate Resolution Imaging Spectroradiometer (MODIS) data and Landsat-8 data for rice crop mapping in western Taiwan. Images of MODIS and Landsat-8 taken in 2013 were processed through five main steps: (1) data preprocessing to account for geometric and radiometric errors of Landsat-8 data, (2) MODIS-Landsat data fusion using the spatial-temporal adaptive reflectance fusion model (STARFM), (3) construction of the smoothed time-series perpendicular vegetation index (NDVI), (4) image classification with the fusion data for estimating rice crop area, and (5) accuracy assessment. The comparisons between mapping results and ground reference data indicated satisfactory overall accuracies and Kappa coefficients. This study demonstrates the applicability of MODIS-Landsat data fusion with STARFM in rice crop monitoring.