Establishment of land cover map using object-oriented classification method for VNREDSat-1 data

Abstract

Land cover/land use classification using high resolution remote sensing data has the biggest challenge is how to distinguish object classes from different spectral values, structures, shapes, and spatial elements. This paper reveals the object-oriented classification method to establish the land cover map using VNREDSat-1 data, with a spatial resolution of 10 m. Land cover/land use system is classified according to CORINE with level 3 with 14 types of land cover/land use. Extraction of 14 types of land cover/land use using object-oriented classification method based on reflectance spectral characteristics, shape index, location of objects, brightness, NDVI plant index, and density objects. The overall accuracy of classification result is about 0.71%