Implementation of Smoothing Image Filtering in the Residue Number System
Nikolay Ivanovich Chervyakov, Pavel Alekseyevich Lyakhov, Nikolay Nikolaevich Nagornov, Dmitrii Kaplun, Alexander S. Voznesensky, Danil V. Bogayevskiy · 2019
In this paper, we propose a new method for image smoothing using the Residue Number System (RNS). The essence of the approach is to replace the computationally complex division operation in RNS with the multiplication of all fractional numbers by a power of two and rounding off. As a result of these actions, all subsequent calculations are performed only in fixed-point numbers. Theoretical and practical studies have shown that when a certain accuracy of calculations is achieved, the error that occurs during rounding does not significantly affect the result of filtering an image. It opens the possibility of effective hardware implementation on FPGA and ASIC.