Platform of quality evaluation system for multimedia video communication based NS2

Yu, Guang-hua, Xu, Jie and Cheng, Xiaochun (2018) Platform of quality evaluation system for multimedia video communication based NS2. Journal of Ambient Intelligence and Humanized Computing . pp. 1-12. ISSN 1868-5137 (Published online first)

Full text is not in this repository.

Abstract

The obtained video sequence scores are statistically averaged as the final quality assessment, and the resulting quality assessment results are less accurate. Therefore, a multimedia video communication quality assessment system platform based NS2 is designed. On the basis of compressed sensing theory, image/video signal is the main research object, aiming at the requirements of quality assessment methods in video transmission systems and the determination of observations in transmission, a video signal recovery quality based on redundant observations is proposed. The purpose of the evaluation method is to solve the problem of quality estimation after video recovery for non-structural dimensionality reduction signals in the transmission of video observations. According to the obtained video quality information, an adaptive adjustment scheme of quality information feedback observation rate suitable for video transmission system is designed to solve the problem of determining the observing rate of unknown sparse signal and improve the quality of video recovery. Under the scheme of observation rate adaptive adjustment, the quality assessment model is given considering the requirements of compressive sensing video features and the subjective perceived quality characteristics in the video system. Meanwhile, a video transmission quality evaluation platform is built based on the evaluation model. The experimental results show that the minimum root mean square error of the proposed method is 0.08. The mean square error of traditional methods are much higher than the proposed method, indicating that the proposed method can better simulate the subjective feelings of the human eye, and the obtained quality assessment results are more accurate.

Item Type: Article
Additional Information: ESSN: 1868-5145
Research Areas: A. > School of Science and Technology
Item ID: 25835
Useful Links:
Depositing User: Xiaochun Cheng
Date Deposited: 21 Jan 2019 10:15
Last Modified: 19 Mar 2019 17:54
URI: https://eprints.mdx.ac.uk/id/eprint/25835

Actions (login required)

Edit Item Edit Item