Knowledge Discovery of Remote Sensing Classification Rules Based on Variable Precision Rough Set
Xin Pan, Shuqing Zhang · 2009
Nowadays the rough set method is receiving increasing attention in remote sensing classification; one of the major drawbacks of the method is that it is too sensitive to the spectral confusion between-class and spectral variation within-class. In this paper a novel remote sensing classification approach based on variable precision rough sets (VPRS) is proposed by relaxing subset operators through the inclusion error ß. The new method proposed here is tested with Landsat-5 TM data. The experiment shows that admitting various inclusion errors ß, can improve classification performance including feature selection and generalization ability. The inclusion of ß also prevents the overfitting to the training data.