Perceptual contrast sensitivity based video quality assessment in DCT domain
Accurate assessment of video quality plays an important role in the chain of digital video services, as unavoidable distortions often occur, e.g., transmission errors over error-prone networks.
As humans are the eventual consumers of video services, it is essential to take into account the characteristics of human vision system (HVS) when developing video quality metrics. Sufficient research interests have been attracted in this field and many video quality metrics have been proposed by considering the psychological mechanisms of human perception . As a practical solution, video features characterizing quality distortions can be extracted from degraded video and used to assess how much impact the distortions have on the perceived quality. Pinson and Wolf  have proposed a video quality metric (VQM) based on several features representing the degradations on spatial gradient, chrominance, contrast and temporal information. As motion is the most important information in the temporal domain of video sequences, a motion-based video integrity evaluation (MOVIE) has been proposed to predict video quality along motion trajectories .