Content-based image retrieval using optimal feature combination and relevance feedback

Abstract

With the rapid development of the multimedia technology and Internet, content-based image retrieval (CBIR) has become an active research field at present. Many researches have been done on visual features and their combinations for CBIR, but few on the performance comparison of different visual feature combinations. Therefore, in the paper, different visual feature combinations are firstly compared in retrieval experiments. Moreover, only using low-level features for CBIR cannot achieve a satisfactory measurement performance, since the user's high-level semantics cannot be easily expressed by low-level features. In order to narrow the gap between user query concept and low-level features in CBIR, a multi-round relevance feedback (RF) strategy based on both support vector machine (SVM) and feature similarity is adopted to meet the user's requirement. The experiment results showed that this SVM and feature similarity based relevance feedback using best feature combination can greatly improve the retrieval precision with the number of feedback increasing.