WITHDRAWN: Contrasting the performance metrics of discrete transformations on digital image steganography using artificial intelligence

Abstract

Image steganography is the impressive art of hiding or embedding a secret image within cover image without any loss when subject to human vision system. The digital image steganography is initially performed by using Least Significant Bit (LSB) technique in spatial domain. Later, the image steganography is implemented by using various frequency domain techniques in which the secret images are transformed to frequency domain and their payload bits are embedded to the frequency domain of the cover image. In this work, the most commonly used transformation techniques like Discrete Fourier Transform (DFT), Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) are applied to the colour images of different formats such as portable network graphics (png), bitmap (bmp), joint photographic experts group (jpeg) and tagged image file format (tiff) and their performance metrics are compared. Additional to the various transformation techniques and colour image formats, various payloads of the secret images are also used. From the overall experimental results and the comparison of performance metrics like mean square error (MSE), peak signal to noise ratio (PSNR), structural similarity index (SSIM) and structural dissimilarity (DSSIM), it is evident that DWT is highly recommendable transformation technique for image steganography compared to other techniques expect for lesser payloads, where DFT exhibits better PSNR and for JPEG images, DCT shows better PSNR.