DocumentCode :
501234
Title :
Real-time Multimedia Quality Assessment Based on Media Delay Index and Support Vector Machine
Author :
Rao, Wenbi ; Li, Yingshu
Author_Institution :
Dept. of Comput. Sci. & Technol., Wuhan Univ. of Technol., Wuhan, China
Volume :
1
fYear :
2009
fDate :
6-7 June 2009
Firstpage :
111
Lastpage :
114
Abstract :
The more areas the multimedia used in, the more effective measures about quality of service are required. This paper presents a new method for multimedia quality assessment. This method uses media delivery index algorithm to analyze the situation of current network and get a series of index as media attributes. Using an intelligent algorithm - support vector machine as a classifier which maps the network parameters with end-userspsila feeling. This is a no-reference method to evaluate the media quality. There is no need to access the original signal in the operation phase, and the obtained results correlate well with human perception. And compared with the neural network quality assessment, the new method is more stable and expansive. So this method can be widely used in real-time system to calculate the magnitude of quality of service automatically.
Keywords :
multimedia communication; quality of service; real-time systems; support vector machines; end-user feeling; intelligent algorithm; media attributes; media delay index; network parameter; no-reference method; quality of service; real-time multimedia quality assessment; support vector machine; Algorithm design and analysis; Area measurement; Delay; Humans; Intelligent networks; Machine intelligence; Quality assessment; Quality of service; Support vector machine classification; Support vector machines; media delay index (MDI); object quality evaluation; quality-affecting parameters; support vector machine (SVM); vedio clips;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Natural Computing, 2009. CINC '09. International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-0-7695-3645-3
Type :
conf
DOI :
10.1109/CINC.2009.28
Filename :
5231366
Link To Document :
بازگشت