Multi-feature image relevance feedback retrieval based on color and texture

A new image relevance feedback retrieval method based on multi-feature is presented. The circular region F-norms in low frequency band of wavelet transform is used as color feature of the image and the energy ratio in high frequency is used to form texture descriptor. In matching the similarity of the images, the Guassian model is used to normalize the different distances between sub-character. Furthermore, relevance feedback technique is used to adjust the weight of every feature in order to satisfy, the user's retrieval goal. Experimental results suggest that the proposed method with low-dimensional features and fewer computation merits improves retrieval performance. It is very suitable for large image database retrieval.