摘要
图像质量评价算法是图像处理研究中的基本问题.深入讨论图像奇异值分解的性质及基于结构相似性的图像质量评价框架,从理论和实验两方面指出两类算法存在错误评价的问题,并从结构相似的观点出发解释了基于奇异值分解的图像质量评价算法,提出了改进算法的思路.
Image quality assessment is a fundamental problem in the field of image processing. Singular value decomposi- tion properties for images and structural similarity-based image quality assessment are deeply discussed. According to both the theoretical and empirical analysis, the drawbacks of current two categories of algorithms are pointed out. Image quality assessment algorithms that based on singular value decomposition are explained from the perspective of the struc- tural similarity. In addition, possible improvement strategies for the current methods are also discussed.
出处
《南京师大学报（自然科学版）》
CAS
CSCD
北大核心
2017年第1期73-78,共6页
Journal of Nanjing Normal University(Natural Science Edition)
基金
江苏省高校自然科学研究面上项目(14KJD110004)
关键词
图像质量评价
奇异值分解
结构相似
image quality assessment, singular value decomposition, structural similarity
