A parallel recurrence method for the fast computation of Zernike moments

This paper presents a parallel recursive method for the computation of Zernike moments from a digital image. The proposed method can reduce the computational complexity of the Zernike radial polynomials by introducing a novel recurrence relation, and be applicable to either the computation of a single Zernike moment or entire set of Zernike moments. The fast computation is achieved because it involves less addition and multiplication operations and is executed in parallel. Moreover, the single Zernike moment can be obtained with employing partial Zernike moments of lower orders. The experiments are carried out to evaluate the performance of the proposed method using binary and grayscale images. The experimental results show that the proposed method takes the shortest time in computing the Zernike moments of a specific order ⩽28 as well as the entire Zernike moments of orders ⩽70.