A robust video watermarking using simulated block based spatial domain technique

Abstract

A digital watermark embeds an imperceptible signal into data such as audio, video and images, for different purposes including authentication and tamper detection. Tamper detection techniques for video watermarking play a major role of forensic evidence in court. The existing techniques for concealing information in the multimedia host are mostly based on spatial domain rather than frequency domain. The spatial domain techniques are not as robust as frequency domain techniques. In order to improve the robustness of spatial domain, a watermark can be embedded several times repeatedly. In order for spatial domain techniques to be more efficient, more payload is needed to embed additional information. The additional information would include the redundant watermarks to ensure the achievable robustness and more metadata of pixels to ensure achievable efficiency to detect more attacks. All these required additional information will degrade the imperceptibility. This research focuses on video watermarking, particularly with respect to Audio Video Interleaved (AVI) form of video file format. The block-wise method is used to determine which block exactly altered. A high imperceptible and efficient tamper detection watermarking technique is proposed which embeds in first and second Least Significant Bits (LSB). The proposed technique divides the video stream to 2*2 nonoverlapping simulated blocks. Nine common attacks to video have been applied to the proposed technique. An imperceptible and efficient tamper detection technique with a novel method of video segmentation to comprise more pixels watermarked is proposed. Experimental results show the technique is able to detect the attacks with the average of Peak Signal-to-Noise Ratio (PSNR) as 47.87dB. The results illustrate the proposed technique improves imperceptibility and efficiency of tamper detection.