CLASSIFICATION OF CROPLAND USING ADEOS-II/GLI GLOBAL MOSAIC DATA

摘要

We attempt to detect croplands using ADEOS-II GLI global mosaic data sets alone. Although we could only obtain GLI data sets for about 7 months, they may give much useful information to exam ine each feature of land cover type. We analyzed these data s ets using the UPDM which is suitable for sensor independe nt analysis method, and exam ined four UPDM coefficients and the MVIUPD values as the vegetation index. Our land cover classification system has 16 class items, seven items of them are croplands. The results of global land cover classification are roughly in agreement with products of GLC2000 except for the deciduous forest and the cropland summer closed.