SKINHIDE: A Biometric Inspired High Fidelity Steganography Technique

The steganography is an art and science of hiding data into innocuous objects such that the existence of the hidden data remains imperceptible to an adversary. Steganography in images have varied techniques of implementation and each of them is unique in terms of the data hiding mechanism. It is evident that certain areas in an image are more efficient for hiding data than the other parts of the image. These are called Regions of Interest or ROIs. Human biometrics such as eyes and flesh areas are some of the ROIs that can be used for steganography purposes. This paper presents a biometric based image steganography mechanism in the spatial domain that finds skin tone pixels from an image and identifies the ROI using matrix encoding mechanism with minimal embedding impact. Efforts have been given to ensure that the proposed algorithm conforms to high Imperceptibility and Fidelity which are the primary quality requirements for any image steganography system.