Fusion of Sentinel-1a and Landsat-8 images for improving land use/land cover classification in Songkla Province, Thailand

Abstract

The objective of this study is to compare the performance of different data fusion techniques for improving the land use/land cover types classification accuracy in Hat Yai district, Songkla province, Thailand. SAR Sentinel-1A and optical Landsat-8 satellites are used as standalone inputs as well as to perform a data fusion based on the resolution merge and LMVM techniques. The four input datasets are classified with a supervised maximum likelihood algorithm and compared against base land cover maps; the results indicate that resolution merge of optical and SAR satellite images can significantly improve the interpretation and classification accuracy of land cover and land use types at the area of interest.