A Video Quality Assessment Metric Based on Human Visual System

To improve video encoding efficiency and performance of video signal processing,an objective video quality assessment metric based on human visual system（HVS）is proposed in this paper by considering the perceptual characteristics to different content,masking effect and spatiotemporal human visual characteristics.Firstly,according to the video spatio-temporal characteristics extracted by three-dimensional Sobel matrix,the proposed metric divides video frame into three regions,and extracts region of interest in combination with just noticeable difference and absolute difference maps.Secondly,the perceptual weighting is used in the region of interest,and the linear weighting is used for spatial domain pooling.Finally,in consideration of HVS’s temporal characteristics,the proposed metric uses low-pass function to simulate HVS’s asymmetric perceptual characteristics and uses worst-case pooling strategy to obtain the final objective evaluation value of video sequence.The proposed metric is tested on the LIVE subjective quality video database,and the experimental results show that the proposed metric can perform better and its evaluation results are more consistent with subjective perception,compared with other classical methods.