An Effective approach for Lossless and High Capacity Image Steganography

This paper investigates the problem of image steganography. The existing algorithms provide low performance due to disregarding the structure and properties of images. Furthermore, most of these schemes are lossy. Considering the shortcomings of the previous algorithms, we propose an effective algorithm, namely Lossless and High Capacity Image Steganography (LHCIS), which embeds the secret image into the cover image losslessly. The proposed scheme comprises secret image compression and embedding the compressed image into the cover image using the LSB algorithm. The compression procedure includes two phases. The first phase constructs the corresponding difference image to the given secret image. The mentioned image represents the variations of the adjacent pixels of the secret one. Each pixel of this image is derived by subtracting its corresponding pixel in the secret image from the adjacent one. The difference image is encoded to reduce its storage space in the second phase. The compression ratio is high due to the small values of the pixels of the difference image. According to the performed simulations, LHCIS increases the embedding capacity by 40%. Furthermore, it improves the PSNR of the stego image by 2.16 in comparison to the previous algorithms.