An edge-based attribute filter dedicated to image segmentation

From LRDE

Abstract

We propose a new edge-based method dedicated to image segmentation. The last few years have shown the development of image processing with connected filters. Indeedcontour-preserving properties of connected operators are highly desired for image segmentation. The connected operators-based segmentation methods usually proceed in two steps. They first compute an attribute on the connected components and filter the components of which attribute do not satisfy a criterion. In these methods, attributes are actually computed on the pixels of connected components. We propose a new union-find based algorithm that enables evaluation of attributes on connected component edges so that we can finally compute an attribute on their contours. We therefore introduce edges-based attributes and propose some of them that evaluate a connected components energy. We finally perform an edge-based attribute filtering to produce a new image segmentation method.