Fusion of colour and texture features in image segmentation: an empirical study

Abstract A study on the fusion of colour and texture features in image segmentation is presented. A series of experiments is conducted to analyse and compare the efficacy of various indexing approaches at every stage of the selection of colour and texture features. The fuzzy c-means clustering algorithm is then applied to combine the selected colour and texture features for image segmentation. The experiments demonstrate that the proposed approach is effective and is able to achieve favourable results in terms of precision and recall when compared with those from the global and local histogram methods.