The Open Automation and Control Systems Journal

2014, 6 : 1444-1450
Published online 2014 December 31. DOI: 10.2174/1874444301406011444
Publisher ID: TOAUTOCJ-6-1444

Research on No-reference Video Quality Assessment Based on H.264/AVC Bitstream

Zhou Zheng and Feng Weidong
Information and Telecommunications Branch, Hubei Provincial Electric Power Company of State Grid, Hubei Wuhan 430077.

ABSTRACT

It is an important part to efficiently evaluate the video quality for user experience in video systems. Traditional video evaluation criterion is strict for testing environment, and is not appropriate for real-time evaluation on network video. No-reference video quality assessment doesn’t require any information of original video, which makes it meet the characteristics of high real-time evaluation on network video image quality and small resource reduction. By extracting key parameters distorted by characteristic video in H.264/AVC video stream flow including quantification parameter, encoding rate and bits of each macro block of coding video, the paper establishes no-reference video quality assessment model from the space complexity and time complexity of video. For the model, not only the model is not complicated, and can meet the requirement of real-time quality assessment of video, the simulation test but also proves that the correlation between the algorithm and subjective assessment value is high, which meets visual perception feature of human eye.

Keywords:

Bitstream , H.264/AVC, no-reference, video quality assessment.