Abstract

How to retrieval efficiently similar images from large images database is a great challenge for content-based image retrieval system. A modified K-means cluster algorithm is proposed to form the hierarchy of the indexing structure. A search method based on A*, triangle inequality principle and N-nearest neighbours is applied to achieve an optimal search for the large image database. Experiments on Corel database are carried out that the proposed method based on the hierarchical indexing structure and new search algorithm achieves the logarithm retrieval efficiency.

Paper Details