Image content adaptive color breakup index for field sequential color displays using a dominant visual saliency method

Zong Qin*, Ying Ju Lin, Fang Cheng Lin, Chia Wei Kuo, Ching Huan Lin, Norio Sugiura, Han Ping David Shieh, Yi-Pai Huang

*Corresponding author for this work

Research output: Contribution to journalArticle

5 Scopus citations

Abstract

An index that can predict the perceptual visibility of color breakup for varying image content is valuable in field sequential color displays, whereas the current indices are usually for fixed patterns. To solve this problem, an image database containing 25 diverse reference images and 125 test cases with various color breakup visibility was first established. Next, visual experiments using a 240-Hz liquid crystal display were performed to acquire the subjective color breakup scores of the test cases. A theorem based on visual saliency theory was proposed that the color breakup perception is mainly determined by the image regions with visual saliency values higher than a certain threshold, called the dominant visual saliency regions. A computational model based on this theorem was developed to obtain objective color breakup scores of the test cases from retinal images with and without color breakup. An analysis of the objective and subjective results revealed a Pearson linear correlation coefficient as high as 0.82, which matches the top-level image quality assessment algorithms. Finally, the proposed color breakup index was used to benchmark against several mainstream field sequential color algorithms to determine their performances in color breakup suppression.

Original languageEnglish
Pages (from-to)85-97
Number of pages13
JournalJournal of the Society for Information Display
Volume26
Issue number2
DOIs
StatePublished - 1 Feb 2018

Keywords

  • color breakup
  • field sequential color display
  • image quality assessment
  • visual saliency

Fingerprint Dive into the research topics of 'Image content adaptive color breakup index for field sequential color displays using a dominant visual saliency method'. Together they form a unique fingerprint.

  • Cite this