【摘要】 准确的客观视频质量评价方法对于视频应用发展是至关重要的.近年来,图像质量评价方法已经比较成熟,而视频质量评价方法与图像质量评价方法在性能上的差距仍然较大.本文提出一种基于四元数奇异值分解的客观视频质量评价方法,该方法将像素的亮度、色度、边缘能量和残差能量作为四元数的四个部分,并用熵作为视觉感兴趣系数对块加权.在视频质量专家组(VQEG)阶段IFR-TV视频测试库上的测试结果表明,该方法与主观评价结果有很好的相关性.更多还原

【Abstract】 Accurate objective quality metric is crucial for various video processing applications.Although the area of image quality assessment has attained maturity in recent years,video quality assessment still has a long way to go before it reaches the levels of success achieved by still image quality metrics.In the paper,we present a new approach to video quality assessment based quaternion singular value decomposition(QSVD),which take the luminance,chrominance,edge energy and motion energy of each pixel as the four parts of a quaternion,and the entropy as the degree of visual interest of regions to weight each block.We demonstrate the performance of our algorithm by testing it on the Video Quality Expert Group(VQEG)Phase I FR-TV dataset,the results show that our method is competitive with state of the art quality assessment methods and has good correlation with perceived video quality.更多还原