Abstract: A video quality assessment model is proposed to improve the correlation with opinion score. Based on the human visual with different sensitivities in the frequency domain, with approximation coefficients and detail coefficients of wavelet transform, low frequency error and high frequency error of frame are calculated by different methods. The product of two errors is adopted as frame objective evaluation of quality. The overall quality of the entire video is given by the average of all frames measurement. Experimental results show that the method is competitive with state of the art quality assessment methods and has good correlation with perceived video quality.

Key words: Video Quality Assessment(VQA), wavelet transform, Sense Capacity(SC), Approximation Coefficient Error(ACE), Detail Coefficient Error(DCE)

摘要： 为提高客观视频质量评价算法与主观评价间的相关性，提出一种基于小波的视频质量评价模型。根据人眼对图像高低频失真的敏感度不同，利用小波变换的近似系数和细节系数，采用不同方法计算各帧图像的高频和低频失真量，将2种失真量相乘得到帧度量，所有帧结果的平均值作为客观视频质量评价值。实验结果表明，该模型结果与主观评价结果具有较好的相关性。

关键词: 视频质量评价, 小波变换, 感觉容量, 近似系数误差, 细节系数误差

CLC Number: 

• TP37