Combining frequency and spatial domain information for fast interactive image noise removal

Anil N. Hirani, Takashi Totsuka

Research output: Contribution to conferencePaperpeer-review

Abstract

Scratches on old films must be removed since these are more noticeable on higher definition and digital televisions. Image noise removal is an image restoration process which replaces the noisy pixels by a prototype image using a fast iterative algorithm which use either frequency domain information and spatial domain information. Existing noise removal algorithms have difficulty with noise which consists of many contiguous pixels and is in a texture area of image or areas with prominent lines. The pixels determined by the algorithm to replace the noise are as sharp as the surrounding area; maintain continuity of prominent lines running across the noise pixels and have texture matching the surrounding texture.

Original languageEnglish (US)
Pages269-276
Number of pages8
StatePublished - Dec 1 1996
Externally publishedYes
EventProceedings of the 1996 Computer Graphics Conference, SIGGRAPH - New Orleans, LA, USA
Duration: Aug 4 1996Aug 9 1996

Other

OtherProceedings of the 1996 Computer Graphics Conference, SIGGRAPH
CityNew Orleans, LA, USA
Period8/4/968/9/96

ASJC Scopus subject areas

  • Computer Science(all)

Fingerprint Dive into the research topics of 'Combining frequency and spatial domain information for fast interactive image noise removal'. Together they form a unique fingerprint.

Cite this