Design of Gaussian star filter for reduction of periodic noise and quasi-periodic noise in gray level images

This paper proposes a new filtering method for periodic noise reduction in gray level images. The proposed filter consists of two orthogonal Gaussian filters with elliptic profile for each noise peak provided a star shaped filter. Therefore it is called “Gaussian star filter” (GaSF). The filter parameters for each noise peak are estimated by region growing algorithm in image amplitude spectrum. Obtained results show that the periodic noise and quasi periodic noise are effectively reduced by the GaSF filtering method.