Colour image filtering is not an obvious task when considering rank order filters. In this paper, an original way of setting up a spatially adaptive median filter is described. The proposed methodology is based on the computation of a colour distance map. Such a map allows the estimation of the optimal width of the filtering window at each point of the image to process. The sort of colour vectors, inherent to a median filtering approach, is achieved by using a bit-mixing paradigm. Finally, experimental results reported in this paper show that the proposed method is able to remove noise whereas fine details and edges are preserved. At the same time, the method is computationally efficient and very easy to implement.