Abstract
Polar harmonic transforms (PHTs) are superior to Zernike moments and Pseudo-Zernike moments in terms of higher speed and numerical stability. Despite all these advantages, there is still a need for fast computing algorithms for real-world applications. So far, the approaches used in boosting the computational speed of PHTs include recursive relation and symmetric/anti-symmetric properties of kernel functions. Taking advantage of these two approaches, a new class of computational framework has been presented to compute the kernel functions with a minimal number of arithmetic operations. The proposed computational framework reduces the number of additions/subtractions from 56 to 24 and the number of multiplications from 12 to 8 compared to existing fast methods. The experimental results show that the proposed method is around 1.4 times faster than that of the existing fastest algorithm.