Investigations on use of fractional Fourier transform for image restoration in the Wiener and geometric mean filters
Kamalesh Kumar Sharma, Priya Mittal · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2013
Image restoration using Wiener and geometric mean filtering is one of the commonly used techniques in image processing applications. In this paper we propose the use of discrete fractional Fourier transform in place of conventional discrete Fourier transform (DFT) in the Wiener and geometric mean filters. The use of discrete fractional Fourier transform (DFrFT) provides us additional degree of freedom in terms of the angle parameter of the transforms which can be exploited for the purpose of image restoration. The proposed restoration filters are applied on both colored and grey images and the simulation results of the proposed technique are presented. The effect of variation of parameters of the transforms and filters are also studied under the presence of noise. It is observed that the results of the conventional Wiener and geometric mean filters are better than the filters using DFrFT except for a specific value of the angle parameter about 0.8.