Evaluation of similarity measurement for image retrieval

Similarity measurement is one of the key issues in content based image retrieval (CBIR). In CBIR, images are represented as features in the database. Once the features are extracted from the indexed images, the retrieval becomes the measurement of similarity between the features. Many similarity measurements exist. A number of commonly used similarity measurements are described and evaluated in this paper. They are evaluated in a standard shape image database. Results show that city block distance and /spl chi//sup 2/ Statistics measure outperform other distance measure in terms of both retrieval accuracy and retrieval efficiency.