An approach to combining dominant color of partition with relevance feedback in image retrieval

Content-based image retrieval becomes a hot research topic due to the rapid increase of out of order image data. In order to solve the problem of lacking the spatial knowledge of common color features, this paper presents a spatial color indexing approach using the dominant colors of multi-resolution partitions. Meanwhile, the technology of relevance feedback is combined with the presented algorithm to enhance the effectiveness of retrieval. A prototype system, namely SmartEye, was implemented. Also, two experiments were designed to evaluate the effectiveness of the approach. Comparison of the retrieval results of SmartEye to that of other systems shows that the proposed approach is effective.