Abstract: A method based on color cluster is used to divide images into regions，and the Gabor wavelet features and gray-level co-occurrence matrix texture features of each region are extracted，and information entropy is used to select feature and reduce the feature dimensionality，then using these features to cluster image region and the semantic feature is gained；then using genetic fuzzy C-means clustering algorithm to cluster image.In image retrieval，the querying image is compared to cluster center，then retrieved in the class with the minimal distance.The experiment results show that the proposed approach has an excellent retrieval precision.