Key technique research of image retrieval based on combination feature

In this paper, we introduce the generalization of feature-based image retrieval, the technology of semantic query and how to use CBIR on web. It research the different methods retrieval the image, including color-shape feature (CS) and color-texture-shape (CTS) feature. Through contrasting CS, CST we found CS and CST enhanced effectively and decreased the storage space and calculation. On the base of the color, texture and shape features, the technology of semantic query is proposed in this paper. The query not only reflect the low-layer physical property of image, but also joined with the high-layer semantic property of image. The retrieval performance is improved efficiently.