Abstract: Shape feature extraction and description arc one of important research topics in content-based image retricval. The paper presented an approach using Zernike moments with minimum geometric error and numerical integration error,which is used for shape feature extraction and description. The approach maps the region of interest in an image into a unit disk,and the Zernike moments can be formed by computing a projection of the mapped image onto Zernike polynomials. Also the psychophysioiogical research results are introduced in the computation of Zernike moments to improve the retrieval performance of a system. Compared with the traditional Zernike moments, our experiment results show that the Zernike moments with minimum geometric error and numerical integration error arc better than the traditional Zernike moments approaches from the viewpoint of reconstruction. Viewed from retrieval, the systems using the Zernike moments with minimum geometric error and numerical integration error have better retrieval performance than the systems using the traditional Zernike moments.

Key words: Content based image retrieval,Shape feature extraction and description,Zernike moments