关键词: 图像质量评估, 人眼视觉系统, HWD变换, 结构相似度评价方法

Abstract:

In this paper,a metric that can measure image quality across scale,space and orientation (SSO-IQM) is proposed based on the fact that human visual system is highly tuned to the stimuli which are “narrow-band” in frequency,space and orientation.The newly proposed metric compares the structural similarity between the reference and the distorted image in the HWD space,which is then weighted by the correlation map to obtain the local quality measure for each subband.To measure image quality across orientation,the weighting factors of the subbands are assigned different values experimentally.Furthermore,extensive experiments indicate that the proposed metric SSO-IQM is more consistent with the subjective evaluation than all the other competing methods evaluated.

Key words: image quality assessment (IQA);HVS;HWD;SSIM