Abstract: Nowadays, network video is widely deployed everywhere. But the objective quality assessment of network video is a challenge, because it will be distorted by various factors, including transmission and compression. This paper proposes a new objective assessment methodology based on fuzzy inference system of Mamdani. Firstly six quality parameters [initial buffering time (Tinit), mean re-buffering duration (Trebuf), re-buffering frequency (Frebuf), Noise standard deviation (Nsd), Blur degree (Bd), and Block effect (Be)] are introduced, and they are all used as input for the fuzzy logic controller system. Secondly, the outputs are used as inputs to another fuzzy logic controller system to obtain the objective quality of network video. Lastly the proposed method is tested on four videos under different network environment and compared with other methods. The experimental results show that the proposed method can improve the similarity between subjective and objective assessment.

Key words: network video, quality parameter, fuzzy inference system, objective quality assessment