Fast Computation of Zernike Moments For Rice Sorting System

In a rice sorting system, the separation distance of one grain from another is determined by the amount of time taken to process the image of each grain. The closer the separation distance of one grain from another will lead to higher output of grains imaged and processed. Hence, the fast computation of a reliable set of pattern features is required in a rice sorting system. Zernike moment, with its rotational invariant property makes it ideally suited to be used. Many fast computation algorithms to compute Zernike moments have been developed. In this study we propose a new application of the symmetrical property of Zernike polynomials to compute a set of Zernike moments in rice sorting system. The average overall classification time of a rice grain is reduced almost by 25.0% when the proposed technique is integrated with the q-recursive method, which currently uses the minimum time to compute a set of Zernike moments.