Change detection using adaptive fuzzy neural networks: Environmental damage assessment after the Gulf War

This article introduces an adaptive fuzzy neural network classifier for env ironmental change detection and classification applied to monitor landcover changes resulting from the Gulf War. In this study, landcover change is tr eated as a qualitative shift between landcover categories. The Change Detec tion Adaptive Fuzzy (CDAF) network learns fuzzy membership functions for ea ch landcover class present at the first image date based on a sample of the image data. An image from a later date is then classified using this netwo rk to recognize change among familiar classes as well as change to unfamili ar landcover classes. The CDAF network predicts landcover change with 86% a ccuracy representing an improvement over both a standard multidate K-means technique which performed at 70% accuracy and a hybrid approach using a max imum likelihood classifier (MLC)/K-means which achieved 65% accuracy. In th is study, we developed a hybrid classified based on conventional statistica l methods (MLC/K-means classifier) for comparison purposes to help evaluate the performance of the CDAF network. The CDAF compared with existing chang e detection methodology has two features that lead to significant performan ce improvements: 1) new landcover types created by a change event automatic ally lead to the establishment of new landcover categories through an unsup ervised learning strategy, and 2) for each pixel the distribution of fuzzy membership values across possible categories are compared to determine whet her a significant change has occurred. (C)Elsevier Science Inc., 1999.