Analysis of Maximum Likelihood classification technique on Landsat 5 TM satellite data of tropical land covers

Abstract

The aim of this paper is to carry out analysis of Maximum Likelihood (ML) on Landsat 5 TM (Thematic Mapper) satellite data of tropical land covers. ML is a supervised classification method which is based on the Bayes theorem. It makes use of a discriminant function to assign pixel to the class with the highest likelihood. Class mean vector and covariance matrix are the key inputs to the function and can be estimated from the training pixels of a particular class. In this study, we used ML to classify a diverse tropical land covers recorded from Landsat 5 TM satellite. The classification is carefully examined using visual analysis, classification accuracy, band correlation and decision boundary. The results show that the separation between mean of the classes in the decision space is to be the main factor that leads to the high classification accuracy of ML.