No-reference Video Quality Assessment Model Considering The Network Packet Loss

Nowadays, people are becoming more and more like watching high-definition video, whether arts or film and Television. At the same time, are all feel distressed with the network video quality which watch from the network terminal that is the user’s computer can not meet the needs of the users themselves. Therefore, caused the thinking of the majority professionals of the academia and industry, what causes the decrease of the quality of the video and how to evaluate the quality of the network video and how to improve the quality of the network video we watched? As is known to us, the major form of network video transmission distortion is delay, packet loss and jitter. Among them, delay almost has no effect on the quality of the video, while the network jitter caused the effect of the packet loss. Therefore, we study around the packet loss. In the end, this paper based on the study of video quality assessment model considering the network packet loss, to meet the needs of the user’s terminal,that is the user’s quality of experience QoE.This paper is mainly carried out under the funding of the National Natural Science Foundation “IPTV video quality Multiple feature extraction and fusion evaluation model research”, the paper based on investigating the existing video quality assessment method, research on no-reference video quality assessment model considering the network packet loss, the specific research contents and results are as follows:(1)The influencing factors of network packet loss’ s long-range dependence has impacts on the packet loss rate. The main content of this chapter is discuss the number of superimposed source N, shape parameter, Hurst parameter, the output link speed has impacts on long correlation, and further affects the packet loss rate. Finally it is concluded that different output link speeds have a significant impact on long-range dependence, and further to affect the packet loss rate. At the same time,this conclusion will be applied to the third part of the contents, select the output link speed as a parameter to establish the model.(2)Mapping model of packet loss rate and the Quality of experience on the influence of packet loss on QoE. The main content of this chapter is using MPEG4 codec and HD video under MyEvalvid platform research on the influence of packet loss on QoE and set up the mapping model of packet loss rate and the Quality of experience in matlab environment. Finally concluded that the different content complexity and different packet loss rate have a significant impact on the user’s quality of experience. At the same time, this conclusion will be applied to the third part of the contents, select content complexity and packet loss rate as parameters to establish the model.(3)No-reference video quality evaluation model considering the network packet loss.On the basis of the above two contents, the main content of this chapter is based on the method of least squares support vector machine(LS-SVM) to establish the no-reference video quality evaluation model considering network packet loss,considering the main parameters are the output link speeds, different packet loss rate and the video content complexity and quantitative parameters and so on. Finally experiments show that, the LS-SVM has better generalization ability, and the training speed is faster.In this paper, based on the video codec, network video transmission distortion,video quality assessment methods, research results of information fusion, committed to the packet loss has effects on the quality of network video, research to find that the No-reference video quality evaluation model considering the network packet loss.First, study on the influencing factors of network packet loss’ s long-range dependence and their has impacts on the packet loss rate. Second, set the Mapping model of packet loss rate and the Quality of experience on the influence of packet loss on QoE.Third, based on the above two parts, set up the No-reference video quality evaluation model considering the network packet loss. Ultimately allowing users to get a good visual experience, thus in network QoE used to drive the control of resources and services optimization. Through a reasonable set of QoS parameters and codec parameters, not only the user get visual enjoyment, and improve the resource utilization of the network and the service provider’s service efficiency. And is beneficial to the optimization of digital video assessment system, the quality ofcommunication monitoring, development in the field of consumer media grading.