Automated interpretation of sub-pixel water cover from satellite images

Accurate information and estimation of area under water cover is important for different engineering applications and decision processes. This paper defines a method for direct interpretation and estimation of the sub-pixel water cover from satellite images. It is a fuzzy knowledge based approach wherein features are represented by Linguistic Variables. In this study, more than 95% accuracy has been found in case of hard classification and 85% in case of sub-pixel estimation. The proposed approach can be utilised in an automated image understanding system.