In this paper, we explore the possibility to exploit GEOBIA concepts for extracting features from multitemporal SAR images. The proposed processing chain is feed by the recently introduced products of the Level-lα and Level-1β families and aims at providing an unsupervised tool for information extraction particularly oriented toward the end-user community. The principal characteristics and the effectiveness of the framework are illustrated through two examples concerning urban area mapping and small reservoir extraction in semiarid environment.
展开▼