High spatial resolution remote sensing image segmentation using marker based watershed algorithm
B. Mathivanan, Sowmya Selvarajan · 2012
W e propose an edge embedded marker based watershed algorithm for high spatial resolution remote sensing image segmentation. Two improvement techniques are proposed for the two key steps of maker extraction and pixel labeling, to make it more effective and efficient for high spatial resolution i mage segmentation. Moreover, the edge information, detect ed by the edge detector embedded with confidence, is used to direct the two key steps for detecting objects with weak boundary and improving the position al accuracy of the objects boundary. It performs well in retaining the weak boundary and reducing the u ndesired over - segmentation.