“…Researchers have attempted to solve problems associated with classification of multispectral imagery by utilizing Neural networks (e.g. Basse, Omrani, Charif, Gerber, & B odis, 2014;Joshi, Leeuw, Skidmore, Duren, & van Oosten, 2006), Evidential reasoning (Franklin, Peddle, Dechka, & Stenhouse, 2002), Decision trees (Zhai, Sun, Sang, Yang, & Jia, 2012) and expert systems (Smirnov, Levashova, & Shilov, 2015;Weng, 2011), markov random (Li, 2009), hybrid classifier (Were et al, 2013), integration of GIS or other ancillary data (Rozenstein & Karnieli, 2011), image fusion (Ehlers, Klonus, JohanÅstrand, & Rosso, 2010) thereby integrating multisource information in the classification. Lu and Weng (2007) discussed in detail several issues affecting the success of a classification among them: the selected remote sensing data, image processing, classification procedure and complexity of the landscape.…”