USE OF PROBABILITY ENTROPY FOR THE ESTIMATION AND GRAPHICAL REPRESENTATION OF THE ACCURACY OF MAXIMUM-LIKELIHOOD CLASSIFICATIONS

A method is proposed for statistically evaluating the accuracy levels of maximum likelihood classifications and representing them graphicall y. Based on the concept that the heterogeneity of maximum likelihood m embership probabilities can be taken as an indicator of the confidence for the classification, such a parameter is estimated for all pixels as relative probability entropy and represented in a separate channel. After a brief presentation of the statistical basis of the methodolog y, this is applied to a conventional and two modified maximum likeliho od classifications in a case study using Landsat TM scenes. The result s demonstrate the efficiency of the approach and, particularly, its us efulness for operational applications.