Reduced-Reference Image Quality Assessment
Image Quality Assessment (IQA) has been recognized as an effective and efficient way to predict the visual quality of distorted images. Various wavelet transforms based methods are used to extract singularity structures, but they fail to explicitly extract the image geometric information, e.g., lines and curves. In this paper, the authors develop a novel framework for IQA to mimic the Human Visual System (HVS) by incorporating the merits from Multiscale Geometric Analysis (MGA), Contrast Sensitivity Function (CSF), and the Weber's law of Just Noticeable Difference (JND). MGA can be used for decomposition of image and feature extraction. CSF is used to balance the MGA decomposed coefficients via a weighting scheme.