The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Articles | Volume XLII-1/W1
31 May 2017
 | 31 May 2017


W. Pervez, S. A. Khan, E. Hussain, F. Amir, and M. A. Maud

Keywords: Change Detection Analysis, Satellite Image Processing, Remote Sensing, Imaging Sciences, Operational Land Imager

Abstract. This paper investigated the potential utility of Landsat-8 Operational Land Imager (OLI) for change detection analysis and mapping application because of its superior technical design to previous Landsat series. The OLI SVM classified data was successfully classified with regard to all six test classes (i.e., bare land, built-up land, mixed trees, bushes, dam water and channel water). OLI support vector machine (SVM) classified data for the four seasons (i.e., spring, autumn, winter, and summer) was used to change detection results of six cases: (1) winter to spring which resulted reduction in dam water mapping and increases of bushes; (2) winter to summer which resulted reduction in dam water mapping and increase of vegetation; (3) winter to autumn which resulted increase in dam water mapping; (4) spring to summer which resulted reduction of vegetation and shallow water; (5) spring to autumn which resulted decrease of vegetation; and (6) summer to autumn which resulted increase of bushes and vegetation . OLI SVM classified data resulted higher overall accuracy and kappa coefficient and thus found suitable for change detection analysis.