Design and Implementation of Steganographic Engine for Secret Data Sharing and Authentication of Videos_ An Overture
Kousik Dasgupta · University · 2014
Steganography is the act of hiding a message inside a container in such a way that can only be detected by its intended recipient. The container can be of different types text, image, audio and video. With the explosion of Internet and reach multimedia content, use of video as a container has become topic of interest for the research fraternity. Though, with the use of internet it has become more vulnerable to interception by unauthorized people over the world. One of the solutions to overcome these issues is cryptography, but the information secrecy does not exist any more. As hiding data is for confidentiality, the work focuses on the utilization of digital video as cover to hide data and maintaining confidentiality. Digital video used as cover can be said to be collection of frames and audio. Frames obtained can be of different types, I, B and P frames, each of them containing varied range of information. The message can be hidden in both frames and audio parts. In the present context frames are used as container, further each frame can be said to a collection of pixels or information. The techniques used for video steganography can be differentiated as Spatial domain, Transform Domain, Linear Motion estimation etc. In the present work some spatial domain techniques are discussed. Further in spatial domain the information can be embedded in Least Significant Bit (LSB) or Most Significant Bit (MSB) of the pixel of each frame. LSB based techniques for embedding has been reported, as LSB techniques are prone to Steganalysis attack. This issue is taken care by applying a hashing function over the base LSB technique. Any steganographic scheme can be said to be govern by mainly two factors imperceptibility and embedding capacity. Genetic Algorithm (GA) as an optimization tool for providing imperceptibility is reported. Use of GA also allows embedding in higher order bits thus in a way improving the embedding capacity. The reported works are compared with some existing schemes and results reported.