Cultural Relic Image Retrieval Method Based on Features of SIFT

Aiming at the cultural relic image retrieval problems in digital archaeology museum of Northwest University, an image retrieval method based on Scale Invariant Feature Transform (SIFT) features is proposed in this paper. Firstly, SIFT features of cultural relic images are extracted and Principal Component Analysis method is adopted to reduce the dimensionality, then an approximate nearest neighbor search algorithm is employed in SIFT feature points matching, Finally, for further improving the retrieval accuracy, local geometric constraint is used to control false matching feature points. The experimental result on cultural relic image database shows that this method is feasible, and the proposed method improves the performance of our cultural relic image retrieval system.