Blind Image / Video Quality Assessment Based on DCT-Domain Statistics

As the need of quality assessment increases day by day. Here develops an efficient general-purpose blind image/video quality assessment (I/VQA) algorithm using a natural scene statistics (NSS) model of discrete cosine transform (DCT) coefficients. The image/video quality assessment approach relies on a simple Bayesian inference model to predict image/video quality scores from certain extracted features. The features are based on an NSS model of the image/video DCT coefficients. The estimated parameters of the model are utilized to form features that are indicative of perceptual quality. These features are used in a simple Bayesian inference approach to predict quality scores. The resulting algorithm requires minimal training and adopts a simple probabilistic model for score prediction. Given the extracted features from a test image, the quality score that maximizes the probability of the empirically determined inference model is chosen as the predicted quality score of that image. In case of video, frames are extracted from video and for each frame above process are repeated. When tested on the LIVE database, the system correlates highly with human judgments of quality both in case of image and video.