Pixel-based Classification of Forest Area using Multi-source Satellite Images in Central Mongolia

The aim of this research is to conduct a forest resources study in central Mongolia using the integrated optical and SAR datasets. As data sources, microwave Sentinel-1B C-band dual-polarization data and optical Sentinel-2B images are available. After creation of a pseudo color image using a principal component analysis (PCA), a supervised classification is conducted. For the classification of forest types, a support vector machine (SVM) classification method is used. The result has been evaluated using an accuracy assessment technique.