Decision Level Fusion in Urban Region Expansion Fuzzy Change Detectors Using ASTER Images

The aim of this study is firstly taking the advantages of effects of various change detection techniques. These techniques are differentiated by the way they use the dataset which could lead us to gain complementary outputs and ultimately achieve higher accuracy in change detection and secondly comparison of performance of three various group of decision level fusion schemes. The study area is the city of Karaj in Iran. Satellite images used in this study are ASTER images captured on July, 2001 and September, 2012. Change detection have been performed with fuzzy post classification and combined fuzzy spectral–temporal analysis techniques, then results of this techniques fused by means of three various group of decision level fusion schemes included:1) Averaging operators 2) Maximum operator 3) fuzzy integral operators. Results of accuracy assessment that has been done by available land cover maps firstly has shown improvement of change detection accuracy over each single fuzzy change detector and secondly has demonstrated advantage of fuzzy integral methods with respect to two other methods.