Similarity Measurement and Relevance Feedback in Image Retrieval
Ia D · Microcomputer Information · 2007
Based on block histogram algorithm for color image retrieval using in CBIR, a new similarity measure algorithm of the fea- ture vectors is proposed in color RGB three- dimensional space that combines spatial analytic geometry. Meanwhile, the paper im- proves a novel relevance feedback which can adjust region weight in image retrieval. Finally, the experiment results show that the approach proposed in the paper achieves better performance.