Abstract: According to the human visual system (HVS), a color block-histogram image retrieval method based on visual weight is proposed in this paper. First, the image is split into blocks and the color block histograms are formed. Second, the color complexity measure (CCM) of pixels are calculated to reflect the image local variation, and the visual weights for subblock images are calculated. Finally, all color block histograms are weighted by the visual weight, and the similarity between color images is computed by using the weighted color block histogram. Experiments, including comparison with state-of-the-arts, show the effectiveness of our algorithm in improving the retrieval performance.