HE Kun, ZHENG Xiu-qing, JU Sheng-gen, ZHANG Yong-lai. Image Denoising on Improved Total Variation[J]. Journal of University of Electronic Science and Technology of China, 2016, 45(3): 463-468. DOI: 10.3969/j.issn.1001-0548.2016.02.026
Citation: HE Kun, ZHENG Xiu-qing, JU Sheng-gen, ZHANG Yong-lai. Image Denoising on Improved Total Variation[J]. Journal of University of Electronic Science and Technology of China, 2016, 45(3): 463-468. DOI: 10.3969/j.issn.1001-0548.2016.02.026

Image Denoising on Improved Total Variation

  • By analyzing the mechanism of isotropic and anisotropic diffusion in image denoising, a new diffusion function based on the effect of noise on the image pixel variation is designed in this paper. The diffusion performance of this function is isotropic diffusion in the smoothing sub-region and anisotropic diffusion on the edge. Then, an improved total variation denoising model is proposed based on the traditional total variation (TV), and the corresponding discrete iterative function is designed by using the fixed point iteration algorithm. The algorithm uses isotropic diffusion on the smooth region, inheriting the advantages of the isotropic diffusion, reducing the running time of the traditional TV, and protecting the edge by using anisotropic diffusion. Experimental results show that this algorithm can improve image peak signal to noise ratio (PSNR) and visual effects.
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return