Image Smoothing via Truncated Total Variation

Zeyang Dou, Mengnan Song*, Kun Gao, Zeqiang Jiang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

17 Citations (Scopus)

Abstract

We present a new regularizer for image smoothing which is particularly effective for diminishing insignificant details, while preserving salient edges. The proposed regularizer relates in spirit to total variation which penalizes all the gradients, while our method just penalizes part of the gradients and leaves the significant edges unchanged. Though the proposed regularizer is a piecewise function, which is hard to optimize, we can unify it to a mathematically sound penalty. The unified penalty term is easy to optimize using recent fast solvers and hard thresholding operation. We show some potential applications of the proposed regularizer, including texture removal and compression artifact restoration. The results show the efficiency of the proposed regularizer.

Original languageEnglish
Article number8110618
Pages (from-to)27337-27344
Number of pages8
JournalIEEE Access
Volume5
DOIs
Publication statusPublished - 2017

Keywords

  • Image smoothing
  • split Bregman iteration
  • total variation model
  • truncated total variation

Fingerprint

Dive into the research topics of 'Image Smoothing via Truncated Total Variation'. Together they form a unique fingerprint.

Cite this

Dou, Z., Song, M., Gao, K., & Jiang, Z. (2017). Image Smoothing via Truncated Total Variation. IEEE Access, 5, 27337-27344. Article 8110618. https://doi.org/10.1109/ACCESS.2017.2773503