Image restoration via improved Wiener filter applied to optical sparse aperture systems

Image deconvolution problem is a challenging task in the field of image process. In this paper, an improved wiener filtering algorithm is proposed to accurately restore the degraded image of optical sparse aperture systems (OSAS). Differing from Wiener filter which needs to estimate the ratio of power spectrum of noise to power spectrum of original image, our algorithm analyzes the relationship between restored image and original image with the change of this ratio. Based on our algorithm, an optimum restored image in which the ring effect and smoothing effect are able to achieve a perfect balance can be gotten without additional estimation. Moreover, in the case of application to OSAS, several main factors which affect the restored image are expounded. Thus, applying this algorithm to image restoration of OSAS would be of great convenience.