Watershed-based hierarchical SAR image segmentation

Watershed transformation is a powerful image segmentation tool recently dev eloped in mathematical morphology. In order to segment images initially ove rsegmented by watershed transformation, two approaches are considered: one is the thresholding of the gradient image proposed by us which is capable o f keeping more salient image contours; the other is the well known centroid linkage region growing algorithm which merges regions with certain statist ical similarities. By choosing suitable thresholds in the two approaches, h ierarchical image segmentation algorithms can be constructed. A Ratio of Av erages (ROA) edge detector is proposed to replace the morphological edge de tectors prior to watershed transformation when applied to Synthetic Apertur e Radar (SAR) images. Applications to SAR agricultural image segmentation w ith these hierarchical segmentation algorithms are presented. It is demonst rated that the algorithms are efficient in the segmentation of the SAR imag es and appropriate for land use applications when the land cover is made up of individual plots.