基于犹豫模糊的视频质量主观评估

Subjective Evaluation of Video Quality Based on Hesitant Fuzzy

针对当前国内外视频质量的主观评估数据处理主要使用算术平均值和方差方法,处理方式简单化,未考虑到不同个体感知差异的问题,提出基于犹豫模糊的视频质量评估方法。为充分利用移动终端的传感能力和人的主观体验,开发了一个移动视频测试应用程序,收集开放教育中四种典型微课程场景的用户体验数据。测试视频码率分别为700k、300k、100k、50k,编码方式为H.264,帧率均取15帧/秒。经过模糊计算,在其他条件一致情况下,得到移动端320p视频在码率300k时是最佳编码方案,此时既保持了较低码率,又保持了较高图像质量。研究结果表明,所提出的模型能够获得和客观评估参数SSIM较一致的估计结果,算法简单易行,且可获得较好的用户体验。

In view of the current subjective evaluation data processing of video quality at home and abroad,the arithmetic mean and variance methods are mainly used.The processing method is simple and does not take into account the differences in perception of different individuals.A video quality evaluation method based on hesitant fuzzy is proposed.In order to make full use of the sensing capabilities of mobile terminals and human subjective experience,a mobile video test application is developed to collect user experience data of four typical micro-course scenarios in open education.The bit rates of test video are 700k,300k,100k,and 50k respectively,the encoding method is H.264,and the frame rate is 15 frames per second.After fuzzy calculation,in the case of other conditions being consistent,the mobile terminal 320p video is the best encoding scheme at the bit rate of 300k,which not only maintains a lower bit rate,but also maintains a higher image quality.The research results show that the proposed model can obtain the estimation results consistent with the objective evaluation parameter SSIM(Structural Similarity),the algorithm is simple and easy to implement,and can obtain a better user experience.