Abstract

The objective of this study was to determine and compare the principal components for different satellite imagery in the same study area. Five different remote sensing data sources were tested. They are: (a) (i) the moderate resolution satellite images from the Landsat Enhanced Thematic Mapper Plus (ETM+), (ii) the Indian Remote Sensing Satellite (IRS), and (iii) French Satellite Pour l'Observation de la Terre (SPOT) and (b) (iv) high-resolution satellite images from IKONOS and (v) airborne hyperspectral images taken by the Compact Airborne Spectral Imaging system (CASI). Among all the principle components (PCs) for all the datasets, the first three PCs contain most of the variance of the original datasets and all the other PC bands contain noise for both moderate and high-resolution images. From these results, it was concluded that instead of original images the first three PCs could be used for classifications in agricultural and wetland areas.