Comparison of support vector machines and spectral angle mapper algorithms for Alteration mapping in Pariz area, Kerman province

Exploration of mineral resources by using remote sensing techniques is done in different ways. One way is to use alteration mapping. In this study, Spectral angle mapper and Support vector machine algorithms were used to separate alterations of porphyry copper deposit. ASTER is an advanced multispectral imager that covers a wide spectral region from the visible to the thermal infrared with high spatial, spectral and radiometric resolution. In this study Aster data and spectral angle mapper and support vector machine algorithms were used to classify the alteration zones. The support vector machine (SVM) is a supervised classification increasingly used in geological mapping. It aims to assign the pixels of the image to classes through an optimal hyperplane and separating them, so that the margin between the two classes becomes maximal. Band ratio and principal component methods used to highlight the alteration zones. Th results of the band ratio method were used as training data and the result of principal component analysis were used as testing data . The similarities of support vector machine and spectral angle mapper algorithm with principal component analysis results are equal to 81% and 89%, respectively.