Image restoration is aimed to recover the original scene from its degraded version. Based on the customary regularization technology and the property of the human visual systems, a new method for image restoration based on adopted regularization technology is presented. This technique is achieved by adopting the parameters of the residual term and the regularization term of each pixel in different part of the image. We can get the parameters of each pixel by computing the local variance at each point. Simulation results demonstrate that the method improves restored image quality and does not slow down the convergence of the evaluation function.