Image deblurring using weighted total variation regularisation for half-quadratic model

It is observed that, in the total variation model for image deblurring, if the regularisation term simply involves the first-order difference, details cannot be satisfactorily restored in a deblurred image. A weighted difference as the total variation regularisation term is considered, and the related half-quadratic model is solved to obtain a deblurring algorithm for saving more details and highlighting the edges of an image. The effectiveness of the proposed algorithm is tested by deblurring experiments.