Object Oriented Steganography using Skin Tone Detection and RSA Encryption Scheme

Steganography is the science that involves communicating secret data in an appropriate multimedia carrier, e. g. , image, audio and video files. While cryptography scrambles the secret message to render it unintelligible, on the other hand Steganography hides the very existence of secret data. Thus Steganography does not replace cryptography but rather boosts the security using its obscurity features. Its main objectives are undetectability, robustness, resistance to various image processing methods, compression, and payload capacity. These unique features separate it from other related techniques such as watermarking and cryptography. The proposed work investigates current state-of-the-art methods and provides a new and efficient approach to digital image steganography. It also establishes a robust steganographic system called Steganoflage. Steganoflage advocates an object-oriented approach in which skin-tone detected areas in the image are selected for embedding where possible. The key objectives of the work are: 1) RSA and OAEP Encryption of the secret text message, 2) HSV model based skin-tone detection algorithm using wavelet domain. Each of these components is tested against relevant performance measurements. The results are promising and provide satisfactory PSNR against various image manipulations and noises such as Poisson, Gaussian, Speckle, Salt and Pepper, and transformations such as rotation, scaling etc.