New Full-Reference Visual Quality Assessment Based on Human Visual Perception

This paper presents an objective video quality evaluation method for quantifying the subjective quality of digital mobile videos. The proposed method aims to objectify the subjective quality by extracting edge region features and blockiness parameters. To evaluate the proposed algorithm's performance, we carried out subjective video quality tests with the double-stimulus continuous quality scale (DSCQS) method and obtained differential mean opinion score (DMOS) values for 140 video clips (CIF/QCIF). We then compared the proposed method's performance to that of existing methods in terms of DMOS estimation accuracy. Experimental results show that the proposed method is approximately 25% better than the EPSNR method in estimating actual DMOS values.