Soft Color Morphology: A Fuzzy Approach for Multivariate Images

摘要: Mathematical morphology is a framework composed by set of well-known image processing techniques, widely used for binary and grayscale images, but less commonly to process color or multivariate images. In this paper, we generalize fuzzy mathematical images in such way that overcomes the problem defining an appropriate order among colors. We introduce soft erosion dilation, which are foundations rest operators. Besides studying their theoretical properties, analyze behavior compare them with corresponding morphological operators from other frameworks deal The outstands when handling CIEL $${}^*a{}^*b{}^*$$ space, where it guarantees no colors different chromatic values original ones created. prove be easily customizable also highly interpretable. Besides, they fast provide smooth outputs, more visually appealing than crisp transitions provided approaches.