摘要
无参考视频质量评价算法均是针对特定的失真类型,但一个失真视频中往往有多种失真类型存在,仅对某种特定的失真进行评价无法全面反映出视频质量;鉴于此,基于小波变换提出了一种能够对多种失真类型进行综合评价的图像质量评价算法;最后,通过亮度和运动估计帧加权的方式将上述方法扩展到视频质量评价中;实验证明了本方法具有较高的评价准确性。
All current video quality assessment algorithms are designed for specific distortion type,however,there are a variety of distortion types in a distorted video and the assessment only for specific distortion type cannot fully reflect video quality.In this view,based on wavelet transform,a non-reference image quality assessment algorithm for comprehensive evaluation on many distortion types was proposed.Finally,a luminance and motion estimation frame weighted approach was utilized to extend this algorithm to video quality assessment.The experiments show that the proposed algorithm has high assessment accuracy.
出处
《重庆工商大学学报（自然科学版）》
2012年第9期44-50,共7页
Journal of Chongqing Technology and Business University:Natural Science Edition
基金
国家创新基金(11c26215115768)
重庆市重点攻关基金项目(cstc2011ab2064)
关键词
视频质量评价
支持向量机
小波变换
video quality assessment； support vector machine； wavelet transform
