Color blurred image restoration based on multichannel nonlinear diffusion model
Nan Wu, Guodong Wang, Hong Zhang, Guojia Hou, Baoxiang Huang · 2017
The blurring of the image will bring great inconvenience to recognize and analyse the content of them. In order to get rid of the blur, many scholars make a lot of contributions. In order to elevate the effect of deblurring, we propose a non-blind restoration variational model for color image deblurring based on the variational method. The proposed method uses the nonlinear diffusion equation as a regular term in the variational energy equation. Besides, we process the image by extending the single channel to the multi-channel, which does well in smoothing the filter and preserving the edge information and furthermore improving the quality of the restored image. In order to accelerate the speed of calculation, the Split Bregman method is introduced. Multiple experiments verify the validity of the proposed method, and the proposed method is superior to traditional algorithms in both visual and objective evaluation.