An Image Quality Assessment Index Based on Visual Saliency and Gradient Amplitude for Telemedicine Application

This paper proposes an image quality assessment method based on visual saliency and gradient amplitude. Multi-scale decomposition is used, then with the analysis of the image's phase spectrum in frequency domain, the visual saliency map can be obtained. Meanwhile, the gradient amplitude map is obtained through analyzing the image's contrast feature. The weighted products of the visual saliency map and the gradient amplitude map of all scales are synthesized to get the objective quality evaluation value. Many experiments are made on three image quality assessment (IQA) databases which are available to the public, including TID2008, IVC and A57. The experimental results demonstrate that the method proposed in this paper can obtain greater consistency with human subjective evaluation than that of other related image quality assessment indexes.