Data hiding in videos using Background Subtraction

Abstract

Nowadays, Video Steganography has emerged as the popular cover medium for hiding of information because of its storage capacity. The selection of pixels for hiding data is very crucial and should be random for security measures. Therefore, we have explored the natural randomness by using such video sequences which contain dynamic variations as a cover. The embedding is done by measuring motion in the frames. It has been achieved by using background subtraction technique which estimates the background to detect foreground pixels. These foreground pixels are then used for hiding data. The proposed algorithm is tested on public available Star dataset for validation. Experimental results show that the proposed algorithm can be used for hiding data as it provides good storage capacity and video quality.