A Highly Secure Skin Tone Based Optimal Parity Assignment Steganographic Scheme Using Double Density Discrete Wavelet Transform

Abstract: Steganography is the art of hiding the existence of data in another transmission medium i.e. image, audio, video files to achieve secret communication. It does not replace cryptography but rather boosts the security using its obscurity features. The proposed method uses both Cryptography and Steganography to enhance the security of the message. The secret message is first encrypted using RSA algorithm and then randomized using OAEP[1]. This encoded message is then embedded in the bitmap cover image using frequency domain approach. For embedding the encrypted message, initially skin tone regions of the cover image are detected using HSV (Hue, Saturation, Value) model. Thereafter, a region from skin detected area is selected, which is known as the cropped region. In this cropped region secret message is embedded using DD-DWT (Double Density Discrete Wavelet Transform). DD-DWT overcomes the intertwined shortcomings of DWT (like poor directional selectivity, Shift invariance, oscillations and aliasing)[2]. Hence the image obtained after embedding secret message (i.e. Stego image) is far more secure and has an acceptable range of PSNR. The proposed method is much better than the previous works both in terms of PSNR and robustness against various noises (like Poisson, Gaussian, salt and pepper, rotation, translation etc.)

Keywords: Cropping, DD DWT, DWT, Gaussian, HSV, OAEP, Poisson, PSNR, RSA, Skin tone detection, Stego Image.