Objective Quality Assessment For Digital Video Based On Feature Extraction And Structure Distortion

With the development of multimedia communication, lots of video services come to the stage, so it's necessary to find some methods to exactly assess video quality. Digital video quality assessment plays important roles in various video applications, such as video processing, compression and communication.This paper has investigated thoroughly the methods of subjective and objective video quality assessment.Firstly, we introduce the existing research methods and achievements of video quality assessment, and give more detailed analysis of some models of objective video quality assessment. Then, combining human visual feature, we present an objective video quality assessment model based on the features extracted from video and structure distortion measurement. This model not only measures the perceptual effects of a wide range of impairments such as blurring, block artifact, noise, but also reflects video structure distortion. At last, this paper gives the experiment results and performance analysis of the model, the experiment indicates that the model has the good estimating performance and reflects subjective video quality more exactly.