Gradient-related non-photorealistic rendering for high dynamic range images

Jiajun Lu, Fangyan Dong, Kaoru Hirota

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

A non-photorealistic rendering (NPR) method based on elements, usually strokes, is proposed for rendering high dynamic range (HDR) images to mimic the visual perception of human artists and designers. It enables strokes generated in the rendering process to be placed accurately on account of improvements in computing gradient values especially in regions having particularly high or low luminance. Experimental results using a designed pattern show that angles of gradient values obtained from HDR images have a reduction in averaged error of up to 57.5% in comparison to that of conventional digital images. A partial experiment on incorporating HDR images into other NPR styles, such as dithering, shows the wide compatibility of HDR images in providing source information for NPR processes.

Original languageEnglish
Pages (from-to)628-636
Number of pages9
JournalJournal of Advanced Computational Intelligence and Intelligent Informatics
Volume17
Issue number4
DOIs
Publication statusPublished - Jul 2013
Externally publishedYes

Keywords

  • High dynamic range
  • Image processing
  • Non-photorealistic
  • Rendering

Fingerprint

Dive into the research topics of 'Gradient-related non-photorealistic rendering for high dynamic range images'. Together they form a unique fingerprint.

Cite this