A new robust video watermarking algorithm against cropping and rotating attacks

In this paper we propose a new video watermarking algorithm based on block classification and visual cryptography to enhance the robustness and imperceptibility based on spatial domain. A method of selecting frames as host frames is proposed based on shot segmentation which avoid embedding watermark into all frames. The watermark signal is cropped into small watermarks according to number of host frames in each shot. A method of selecting of host coefficients is proposed based on block classification using edge detection in spatial domain. Owner's share is generated based on block classification and sub-watermark bits. The imperceptibility of the watermarked video is improved by this way. The experiment results show that the proposed scheme has a good performance in maintaining realtime performance and resisting frame averaging, frame reduction and geometric attacks.