摘要: A new semi-supervised `growing' Bayesian classifier for small reservoir delineation has been developed and is tested with Radarsat-2 data reservoirs in the semi-arid Upper East Region of Ghana. The reduces confusion error to land-water boundary pixels, can readily be extended auxiliary information a high degree automation. Results indicate that algorithm able delineate open water from SAR imagery different weather environmental conditions. As such, allows remotely sensed operational monitoring storages.