Land Cover Classification of ALOS Data Using Back-propagation Neural Network Models

Abstract

Land cover information is vital for supporting decision concerning the management of the environment and for understanding the causes and trends of human and natural processes on the earth surface. The extraction of useful information (i.e. land cover) from satellite images, by means of classification, is one of the most important technical problems of remote sensing. The aim of this research is to develop land cover classification method of ALOS data using Back-Propagation Neural Network classifier a case study in Bogor Botanical Garden – West Java. The targets of this research are: to investigate the effect of different input parameter to land cover classification result using Back-Propagation Neural Network and to assess accuracy of the classification result generated with different input and to compare accuracies to decide whether one model is superior to another.