To identify landslides for disaster monitoring, FORMOSAT-2 imagery has the advantages of low cost and frequent revisit over any other satellite imagery currently available in Taiwan. However, its four spectral bands are not capable enough to distinguish landslides from other ground cover types, for example, thin rivers. In this study, we attempt to overcome the spectral incapability of FORMOSAT-2 imagery from the standpoint of classification. First, we explore more discriminative features, such as texture and topographical features, in order to improve class separability. Texture features are extracted from the FORMOSAT-2 imagery itself using the log-polar wavelet packet transformation whereas a topographical feature slope is derived from an auxiliary Digital Elevation Model (DEM) dataset. Second, a contextual classifier that combines spectral and spatial information is used since this type of classifier is appropriate for our case of homogeneous object identification. Our results have been validated partially by field investigation on several sites. Experiments show that our approach has given a significant improvement over the spectral approach.