A full-reference image quality metric for objective evaluation in spatial domain
Akshat Jain, Vikrant Bhateja · 2011
This paper presents a new full-reference image quality metric for objective assessment in spatial domain. This metric empirically combines some standard image metrics for assessment of complete visual distortion. The proposed quality metric (Q) is formulated by modeling an image distortion as a combined effect of structural distortion, contrast distortion and edge distortion, which may occur on account of deteriorations due to noise contamination, contrast manipulations, blurring, rotation or compression. As portrayed by the experimental results, this metric provides a more accurate estimation for the above mentioned deteriorations in comparison to conventional MSE (PSNR) and SSIM approaches in terms of performance with reference to subjective assessment. The novelty of the proposed metric is adjudged by its ability to evaluate across different types of distortions under a single metric for image quality assessment.