The Algorithm for Image Deblurring Based on Smoothing Absolute Value Function
A convex smooth deblurring model is presented which combines the edge information of the image and absolute value smoothing technique.Based on the TV and Tikhonov hybrid regu-larization model,the smoothing function is employed to approximate the TV regularization term of the hybrid model,which makes it a smooth and convex optimization problem,and it is solved by the semi-proximal alternating direction method of multipliers(SPADMM)algorithm.Numerical experi-mental results on different fuzzy kernels and Gaussian noise levels show that our algorithm can not only effectively eliminate noise and blur,but also retain the image edge information better.
NETL
NSTL
万方数据
