Fusion of multisensor and multitemporal data in remote-sensing image analysis

The authors address both the classification and the detection of changes in multitemporal and multisensor remote-sensing images. They propose a technique that is based on the compound classification rule for minimum error. The basic idea of such a technique was presented by L. Bruzzone et al. (1997), where it was applied to the detection of changes in images acquired by a single optical sensor. The purpose of the present paper is to present an improved version of the authors' technique and to highlight its potentialities for the analysis of multisensor images by reporting on experiments with a real data set.