Reversible Steganography for RGB Images Using Image Interpolation

D. Shahi, R.S. Vinod Kumar and V.K. Reshma

Improving the embedding capacity and increasing the robustness are the key roles in steganography. Steganography is a method of invisible communication. Data hiding with the use of interpolation gives high embedding capacity and makes the embedding process reversible. A reversible data hiding in RGB images using image interpolation is proposed in this paper. The three Image Interpolation techniques namely High Capacity Reversible Steganography using Multi-layer Embedding (CRS), Image Interpolation based data hiding (IIDH) using Optimal Pixel Adjustment Process (OPAP) and Reversible Image Steganography using Improved CRS (RSIC) techniques are applied in RGB images. For each actual pixel location, the secret bits are concealed in median plane of the interpolated RGB images which is used as cover media. The method used for secret data embedding is XOR operation, which is carried out between the secret bits and pixel value in the median plane of cover image. This makes the embedding process simple and less complex. The reversibility of the proposed technique helps in the precise retrieval of the cover media at the time of secret data extraction. Since the actual pixel vales are not changed due to embedding process, this can be used in military and medical imaging applications. From the experimental results of the three proposed techniques, it is revealed that good image quality and improved robustness are achieved with these techniques. The PSNR is increased by a value of nearly 5dB for RGB images than gray scale images for all the proposed techniques. Also the RGB images show higher immunity towards attacks than gray scale images, when tested with Gaussian and salt & pepper noise.

Volume 12 | Issue 3

Pages: 41-49

DOI: 10.5373/JARDCS/V12I3/20201165