Half Blind Restoration Algorithm of Building Image Based on Total Variation
Qun Zhou · Journal of Information and Computational Science · 2014
Based on restoration theories, this paper mainly studies the blind restoration method of motion-blurred image, including a joint identification method and a priori fuzzy identification method. For the halfblind restoration problem, a total variation model is provided and a alternate iteration method based on split-Bregman algorithm is proposed to solve this model. By this new model, the parameters in PSF and the real image can be estimated simultaneously. Our method is demonstrated effective with numerical simulations by the restoration of a motion-blurred building image.