Digital Image Steganography Based on Integer Wavelet Transform and Assignment Algorithm

Abstract

In this paper, we present a new image steganography technique based on Integer Wavelet Transform (IWT) and Munkres' assignment algorithm which embeds secret image in frequency domain of cover image with high matching quality. IWT is used to transform both cover and secret images from spatial domain to frequency domain, and assignment algorithm is used for best matching between blocks for embedding. We embed the secret image in different coefficients of cover image bands such as horizontal detail, vertical detail and diagonal detail and observe the effect of embedding on the performance of stego image in terms of Peak Signal to Noise Ratio (PSNR). Experimental results depict that stego image and extracted secret image could have high visual quality and they are perceptually similar to their original versions. In addition, this method shows high robustness against six different attacks. Embedding in diagonal detail coefficients gives better PSNR than other coefficients.