Visual depth guided image rain streaks removal via sparse coding

Duan Yu Chen*, Chien Cheng Chen, Li Wei Kang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

Rain removal from an image is a challenging problem since no motion information can be obtained from successive images. In this work, an input image is first decomposed into low-frequency part and high-frequency part by using guided image filter. So that the rain streaks would be in the high-frequency part with non-rain textures, and then the high-frequency part is decomposed into a 'rain component' and a 'non-rain component' by performing dictionary learning and sparse coding. To separate rain streaks from high-frequency part, a hybrid feature set is exploited which includes histogram of gradient (HoG) and difference of depth (DoD). With the hybrid feature set applied, most rain streaks can be removed; meanwhile, non-rain components can be enhanced. Compared with the state-of-the-art method [12], our proposed approach shows that not only the rain components can be removed more effectively, but also the visual quality of restored images can be improved.

Original languageEnglish
Title of host publicationISPACS 2012 - IEEE International Symposium on Intelligent Signal Processing and Communications Systems
Pages151-156
Number of pages6
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event20th IEEE International Symposium on Intelligent Signal Processing and Communications Systems, ISPACS 2012 - Tamsui, New Taipei City, Taiwan
Duration: 2012 Nov 42012 Nov 7

Publication series

NameISPACS 2012 - IEEE International Symposium on Intelligent Signal Processing and Communications Systems

Other

Other20th IEEE International Symposium on Intelligent Signal Processing and Communications Systems, ISPACS 2012
Country/TerritoryTaiwan
CityTamsui, New Taipei City
Period2012/11/042012/11/07

Keywords

  • dictionary learning
  • difference of depth
  • image decomposition
  • rain removal
  • sparse representation

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Signal Processing

Fingerprint

Dive into the research topics of 'Visual depth guided image rain streaks removal via sparse coding'. Together they form a unique fingerprint.

Cite this