AN INFORMATION FUSION METHOD FOR MULTISPECTRAL IMAGE CLASSIFICATION POSTPROCESSING

Remote-sensing image classification is one of the most important techn iques in understanding the dynamics of the Earth's ecosystems. Various approaches have been proposed for performing this classification task . Obtained classification results are generally shown as a thematic (o r class) map in which each pixel is assigned a class label. Due to sen sor noise and algorithm limitations, obtained thematic maps are very n oisy, The noise has a ''salt-and-pepper'' appearance in homogeneous re gions and produces weakly defined interregion borders. In this paper, a new postprocessing approach aiming to produce thematic maps with sha rp interregion boundaries and homogeneous regions is presented, This a pproach is conducted in two steps: 1) relevant features derived from t he original multispectral image (edge maps) as well as from the themat ic map, the Smoothed Thematic Map (STM), are determined and 2) a regio n-growing algorithm is applied over the thematic map, This algorithm g rows until reaching an edge (from the edge maps) or a class change in the STM, The proposed approach fills the requirements of being indepen dent of the used classification algorithm and not knowledge-based (in the sense that no a priori information concerning the contents of the considered image is needed). Tests have been conducted on a Landsat im age covering mainly agricultural areas.