Automated initialization and automated design of border detection criteria in edge-based image segmentation

An automated model-based image segmentation method is presented. Information for image segmentation is automatically derived from a training set provided in a form of segmentation examples. In the first step, an approximate location of the object of interest is determined. In the second step, accurate border segmentation is performed. The method was tested in five different segmentation tasks that included 489 objects to be segmented. The final segmentation was compared to manually defined borders with good results. Two major problems of current edge-based image segmentation algorithms were addressed: strong dependence on a close-to-target initialization, and necessity for manual redesign of segmentation criteria whenever a new segmentation problem is encountered.