Abstract: A color image retrieval algorithm based on the theory of bit-plane is proposed,which takes into account the color features and the properties of spatial distribution of the color.In order to better describe the retrieval features and to improve the retrieval efficiency,three new concepts are introduced,that is,bit-plane mean,bit-plane flatness and bit-plane roughness.Based on these three concepts,the features and spatial distribution of the color are extracted.Finally,a reasonable similarity measure scheme is designed for these retrieval features.Experimental results show that the proposed image retrieval algorithm is more effective and efficient in retrieving similar images.