Abstract
The subject of this paper ist the semantically baswed postprocessing of color-coded images such as classification results. We outline why the classical definition of mathematical morphology suffers is used for processing of coded image data. Therefore we provide an extension for morphological operations such as dilation, erosion, opening, and closing. With a new understanding of morphology we introduce bridging and tunneling as further combinations of dilation and erosion. The extensions are applied to medical image data, where the semantics rules stem from basic anatomical knowledge.