Monitoring Land Use/Cover Changes Using

Land use/cover change mapping is one of the basic tasks for environmental monitoring and management. In recent years, a variety of change detection techniques have been developed. This research compares three change detection techniques, including image differencing, image rationing, and image regression to study land use/cover changes in Falavarjan /Iran. The data sources used in this study were Landsat Multi-Spectral Scanner (MSS) and AWiFS images taken in September 1972, and September 2008, respectively. First, images were geometrically and radiometrically corrected. The root mean square (RMSe) obtained 0.5 pixels for each images. The Three change detection methods were performed. Then, a supervised maximum likelihood classification was used as a crossclassification to detect “from-to” change which allowed to assess the accuracy of each change detection technique. Based on accuracy assessment, the image differencing method was the most accurate one with an overall accuracy of 85% in detecting land use/cover changes in Falavarjan area. This was followed by the image rationing technique with an accuracy of 84%.