Using Change Detection and Classification Methods to Study the Al-ahwaz Marshal using Landsat-7 ETM Images

Abstract

In this paper, six bands data sets taken from the landsat-7 ETM (Enhanced Thematic Mapper ) to determine the changes which occur in Al-Ahwaz marshal during 2000 and 2013. Different methods were used to study the changes which occur in this region the subtraction method which Image Difference (ID) and the (PCA) method which is Principal Component Analysis. Campaign the (PCA) with (ID) to produce the best way to detect the changes. The unsupervised classification process which is the modified (K-mean) is used to classify the changes occur in the marshal region and the area surrounding it. This type of classification depend on the color and distance between the classes, a hybrid methods which are the (ID) with the K-mean and PCA of (ID) with the K-mean were implemented to produced the best change detection regions.

Keywords: PCA, Image Subtraction, Modified K-mean