Morphological colour operators in totally ordered lattices based on distances: Application to image filtering, enhancement and analysis

The extension of mathematical morphology operators to multi-valued functions, and in particular to colour images, is neither direct nor general. In this paper, a generalisation of distance-based and lexicographical-based approaches is proposed, allowing the extension of morphological operators to colour images for any colour representation (e.g., RGB, LSH and L∗a∗b∗) and for any metric distance to a reference colour. The performance of the introduced operators is illustrated by means of different applications: colour feature extraction using openings (closings) by reconstruction, colour gradients for segmenting, colour denoising by the centre operator and colour enhancement by the contrast mapping. Examples from natural colour images and biomedical microscopic colour images are given.