Realistic Scene Rendition for High Dynamic Range Imaging

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 视频内容安全

	Realistic Scene Rendition for High Dynamic Range Imaging
	Jinhua Wang; De Xu; Bing Li
发表期刊	Chinese Journal of Electronics
	2010
卷号	19 期号:1 页码:75-80
摘要	Currently, most algorithms for High dynamic range (HDR) images focus on dynamic range compression denoted as tone mapping, which can make more details in the dark areas and highlight areas visible. However, the lighting condition in which the images are taken is not considered, as a result, the mapped images may have color cast. We know that Human visual system (HVS) can perceive a scene regardless of the changes in lighting conditions. Therefore, to obtain an image that corresponds to direct observation of the scene for human eyes, dynamic range compression alone is not sufficient. In this paper, we propose a method called Realistic scene rendition (RSR) to deal with HDR images, which is a model combining the dynamic range compression and chrominance adjustment. The two processes are parallel, which is the most difference from other methods that serve the dynamic range compression as a preprocessing. Furthermore, in order to improve the visual quality of images, we propose a contrast enhancement technique. The results show that RSR can maintain details as well as other tone mapping methods and the visual appearance is more similar to the real-world scene perceived by HVS.
关键词	High Dynamic Range Imaging
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/20395
专题	多模态人工智能系统全国重点实验室_视频内容安全
作者单位	Beijing Jiaotong University
推荐引用方式 GB/T 7714	Jinhua Wang,De Xu,Bing Li. Realistic Scene Rendition for High Dynamic Range Imaging[J]. Chinese Journal of Electronics,2010,19(1):75-80.
APA	Jinhua Wang,De Xu,&Bing Li.(2010).Realistic Scene Rendition for High Dynamic Range Imaging.Chinese Journal of Electronics,19(1),75-80.
MLA	Jinhua Wang,et al."Realistic Scene Rendition for High Dynamic Range Imaging".Chinese Journal of Electronics 19.1(2010):75-80.