Image steganographic techniquewith High embedding capacity based on LWT and Modulus Operator

Abstract

A robust steganography scheme must satisfy the fundamental requirements such as the capacity of hiding data in the cover medium, imperceptibility of the hidden data, and security.  There are many schemes/methods have been proposed so far are lacking in anyone of the three. All the schemes are developed either in the spatial domain where the operations are performed directly on cover image pixels or in the frequency domain where the operations are performed on transformed coefficients. Frequency domain schemes are preferred than the spatial domain where the later one is not fulfilling simultaneously both imperceptibility and robustness against steganalysis. Considering all the constraints a new frequency domain scheme is proposed. The proposed method makes use of the power of 2-D lifting wavelet transform (LWT) with the wave “sym3” and modulus operator. The effectiveness of the proposed method/scheme is verified and compared with recent methods/schemes in terms of PSNR and SSIM. Proposed method achieves maximum embedding bit rate of 4 bits/pixel, for instance, an image of size 512x512 can embed 1048576 bits and maintains an average PSNR of 50.5 and average SSIM of 0.99.

Keywords: Lifting wavelet transform (LWT), sym3, Modulus Operator, Stego-Image, Number Conversion.