Information about land covers is essential for a variety of purposes, such as environmental studies, sustainable development, and regional managements. This study aims to use a novel combination model to generate a land cover map in a part of the Cameron Highlands, Malaysia, where there are different kind of land covers, including tea plantation, florification and forest. Because of the high similarity in land covers of the study area, only through satellite imageries with a high spatial and spectral resolution the land covers can be differentiated. We have combined satellite imageries of Sentinel-1 (S1A, GRD, IW) and Landsat-8 (Operational land imager) for the year 2017 as well as different algorithms of Maximum Likelihood (ML), Minimum Distance (MD), Support Vector Machine (SVM), Spectral Angle Mapper (SAM) and Artificial Neural Network (ANN). The results showed that the combination model is an applicable technique for extracting land covers in areas with high similarities in land covers. The overall accuracy of the confusion matrix and the Kappa Coefficient are 98.1984% and 0.9579, respectively, which indicate that it is a robust model for extracting land covers in areas like the Cameron Highlands. The obtained results can be useful for different purposes, including urban and environmental management, change detection, agriculture and many more purposes.
