Uncompressed video quality metric based on watermarking technique

This paper presents a no-reference video quality metric that blindly estimates the quality of a video. The proposed system is based on video watermarking using 8×8 blocks DCT coefficients of YCBCR domain, and for watermark generation; the Geffe generator has been used to generate binary stream sequence watermark in embedding and extracting processor. Data hiding is achieved by simple “even-odd” signaling of the DCT coefficients. The comparison process between the extracted watermark and the generated watermark from Geffe generator was calculated to conclude the video quality assessment by measuring the watermark degradation. An identical watermark within each frame has been used in this system. With these mechanisms, the proposed method is robust against the attacks of frame dropping, averaging, swapping, and statistical analysis. The results indicate that the proposed video quality metric outperforms standard Peak Signal to Noise Ratio (PSNR) and structural similarity and Image Quality (SSIM) metric in estimating the perceived quality of a video.