一种网络丢包的无参考视频质量评估方法

摘要

A novel no-reference video quality assessment model is proposed in order to meet the real-time and low cost requirements of video quality assessment over IP networks. The temporal complexity is calculated without decoding, and the effect of different types of frame loss on video quality is considered. In the proposed model, both spatial and temporal error concealment artifacts of the human visual system (HVS) and frame types are taken into account. Video qualities could be assessed without delay by utilizing the proposed model. Experimental results show that the scores obtained by the proposed model highly agree with the subjective assessment scores. Compared with the video quality evaluation model in G. 1070, an average performance gain of 6. 96 % could be achieved by the proposed model in terms of relativity to subjective results.%针对网络视频质量评估需要满足实时性高、资源开销小等要求,在研究不同类型帧的丢包对视频质量影响的基础上,提出一种不需要对视频解码的无参考网络视频质量评估模型.该方法不必对接收到的码流解码,只通过码流的包头分析即可得到视频的时域复杂度；并通过研究不同帧类型的丢失对视频质量的影响,提出一种无参考的网络视频质量评价模型来预测网络失真视频质量.该方法不仅考虑到人眼视觉系统的空域掩盖效应和时域掩盖效应对视频质量的影响,而且也考虑了不同类型的帧的丢包对视频质量的影响.提出的模型不需要对视频进行完全解码,具有资源开销小、实时性好的特点,适合对实时传输的视频流进行质量评估.实验结果表明,使用该视频质量评估方法测得的视频丢包失真质量评分与其主观质量评分有很好的一致性,相比于国际标准G.1070中的丢包视频质量评价模型,无参考的网络视频质量评价模型与主观质量评价的相关性平均提高了6.96％.