A classification method of pearl shape based on Zernike moment

An approach for pearl shape classification using fuzzy pattern recognition based on Zernike moments was proposed. After a series of preprocessing on the pearl image obtained, transformed it to the polar coordinates, then calculated eighth-derivative Zernike moments means as each type of typical shape ´s characteristic parameter. The utilization of fuzzy pattern recognition achieved the shape effective distinction of pearl single image, finally, by seeking and comparison of characteristics image through multi-angles of view, realized the pearl shape classification. Experiment results indicated that, the method is of good distinction ability.