Regularization Parameter Selection for Total Variation Model Based on Local Spectral Response

In the past decades, various image regularization methods have been introduced. Among them, total variationmodel has drawn much attention for the reason of its low computational complexity and well-understoodmathematical behavior. However, regularization parameter estimation of total variation model is still an openproblem. To deal with this problem, a novel adaptive regularization parameter selection scheme is proposedin this paper, by means of using the local spectral response, which has the capability of locally selecting theregularization parameters in a content-aware way and therefore adaptively adjusting the weights between thetwo terms of the total variation model. Experiment results on simulated and real noisy image show the goodperformance of our proposed method, in visual improvement and peak signal to noise ratio value. KCI Citation Count: 3