智能系统学报2012，Vol.7Issue(5)：414-422,9.DOI:10.3969/j.issn.1673-4785.201206021
基于感知视觉重要性的立体图像质量评价
Quality assessment of stereoscopic images based on the significance of perceptual vision
摘要
Abstract
The visual psychological and physiological factors are crucial for the assessment on image quality. There are many visual, psychological and physiological calculation models used to support the assessment of image quality, however, how to analyze the intrinsic relations among various isolated methods and further effectively coordinate them in the task of image quality assessment is the key to making the assessment results more conform to subjective assessment of human. The definition and formalization of visual attention areas as saliency in the early vision period and poor-quality area for vision transformation periods were examined from an image quality angle. Simultaneously, the property of perceptual redundancy of human eyes, JND(just noticeable distortion) perception redundancy model, and vision perception model PVSSIM of image equality assessment will be examined. Based on the research findings, feasibility of vision perception methods are verified using a two-dimensional image database study through the introduction of image stereoscopic. The experimental results confirm that, in comparison with traditional methods, the raised objective assessment method sufficiently considers the coordination of various psychological and physiological factors in the task of image quality assessment, it is more relevant to the subjective image quality assessment of human. In the stereoscopic image database, the assessment method may properly coincide with the subjective assessment, the expectable effect is realized.关键词
图像质量评价/立体图像评价/JND/视觉心理/生理模型/感知视觉Key words
image quality assessment/ stereoscopic image/ just noticeable distortion(JND)/ visual psychology/ physiological model/ perceptual vision分类
信息技术与安全科学引用本文复制引用
姜峰,郭美思,刘绍辉,赵德斌..基于感知视觉重要性的立体图像质量评价[J].智能系统学报,2012,7(5):414-422,9.基金项目
国家自然科学基金资助项目(61100096) （61100096）
国家"973"计划资助项目(2009CB320905). （2009CB320905）
