On the classification of image regions by colour, texture and shape

Abstract A unified approach to how colour, texture and shape can be encoded in a single spatio-chromatic feature space is developed. In particular, it is shown how such features can be extracted using multi-scaled filtering and correlation methods which capture the variations of colour over space in ways which encode important image features not extracted by techniques which separate colour, texture and shape into separate channels. Finally, it is shown how such spatio-chromatic features can be used in difficult classification problems.