Watershed approaches for color image segmentation

This paper presents and discusses different approaches to extend the watershed transform, which is the classical segmentation tool in gray-scale mathematical morphology, to the case of color or, more generally speaking, multicomponent images. Different strategies are presented and a special attention is paid to the “bit mixing approach”. This method bijectively mapps multi-dimensional data into a mono-dimensional space. It thus allows to use directly the usual scalar watershed algorithm on the encoded data, nevertheless taking explicitely into account their vectorial structure.