Museum Cultural Relics Retrieval System Based on Salient Region

Abstract

In the field of image retrieval, the problem of semantic gap exists in the current content-based image retrieval technology (low-level features can’t describe the high-level semantics accurately and comprehensively), and it is also one of the important problems. At the same time, the museum cultural relic image has its particularity. Search for different databases, many researchers have done a lot of work in image retrieval. We also established a rich cultural relic database and classified the cultural relics. Making use of the SIFT feature extraction method based on saliency and cultural relic image retrieval in experimental museum is improved. It also improves the current cultural relic retrieval performance.