A relevance feedback method in image retrieval by analyzing feedback log file

Content-based image retrieval is one of the most important research areas in multimedia databases. In recent years, it has been more and more popular to use the mechanism of relevance feedback to improve retrieval performance. We propose a novel approach of accelerating the feedback process by analyzing the feedback log file. Experimental results show that our method brings dramatic improvement to retrieval efficiency compared with the traditional relevance feedback method.