Full Reference Quality Assessment of Full HD Images Using Combined Saliency Priors in Multi-Scale

MD, Sameeulla Khan (2018) Full Reference Quality Assessment of Full HD Images Using Combined Saliency Priors in Multi-Scale. In: 24th National Conference on Communications, NCC 2018, 25 to 28 February 2018, Hyderabad; India.

Full text not available from this repository. (Request a copy)

Abstract

Conventional image quality metrics such as SSIM, FSIM etc out performs well in standard definition (SD) and enhanced definition (ED) images. But with the rapid advancement in technology and miniaturization of devices, the present camera devices support high resolution images such as high definition (HD), full HD and ultra HD. Existing 2D IQA methods doesn't perform better over high resolution images, especially over full HD and above resolution images. So we present a saliency based approach whose performance is in par with existing methods but shows state-of-The-method over full HD databases. We also consider gradient magnitudes and inter-gradient maps for predicting structural similarity. The choice of usage of saliency and gradient methods results in best performance over full HD database. Our proposed algorithm also implemented in different scales of image where the scale number depends on the size of the image.

[error in script]

IITH Creators: