Potential of Alos Palsar 50m Mosaic Product for Land Cover Classification in Tropical Rain Forest

In this research, the potential use of 50m ALOS/PALSAR mosaic data in mapping land cover and monitoring deforestation in Riau province using its amplitude and texture is studies. By using several bands of textural information, the main land covers in the image were improved. The geo-statistical analysis indicated that there was no single window size that would adequate characterize the range of texture measures. Classifiers investigated are maximum-likelihood (ML), and Support Vector Machine (SVM). Classification accuracy is measured. The results indicate that SVM with texture gave the best result over the other methods.