A Hybrid Non Blind Video Watermarking Scheme Based on DWPT and PCA

This paper present a novel hybrid watermarking technique which combine two watermarking methods Discrete Wavelet Packet Transform (DWPT) and Principle Component Analysis (PCA). It embed a binary logo watermark into video frames. Combining the two transforms improved the performance of the watermarking algorithm. The performance can be evaluated in terms of the imperceptibility and robustness against various attacks such as contrast adjustment, noise addition and geometric attacks. The Peak-Signal-To-Noise Ratio (PSNR) used to measure the visual quality of watermarked video. NC is the normalized correlation between the original and the extracted watermark. Experimental results has shown no visible difference between the watermarked frames and the original frames, and shown the robustness against a wide range of attacks such as Gaussian noise addition, histogram equalization, salt and paper noise, contrast adjustment, JPEG compression, cropping, resizing, rotation, frame dropping, frame averaging, frame swapping etc.