An Image Steganography Algorithm for Hiding Confidential Data using Genetic Algorithm

Abstract

Steganography is the art of information hiding, which embeds a confidential data into a cover message such as image, audio, video and text. It preserves the existence of a secret data and makes it undetectable. Digital images are commonly used in steganography due to the popularity of digital image transmission via the Internet. However, the tradeoff between high capacity of secret data, and high security and quality of stego image is a key issue. In this thesis, an image steganography technique based on Haar Discrete Wavelet Transform (HDWT), Lempel Ziv Welch (LZW) algorithm, Optimal Pixel Adjustment Process (OPAP) and Genetic Algorithm (GA) is proposed. Firstly, the cover image is divided into non-overlapping blocks of n×n pixels. Then, the HDWT is used to transform each block from a spatial domain to a frequency domain. This increases the robustness of stego image against attacks. To increase the hidden capacity and security, LZW algorithm is applied on the secret message. After that, the GA is employed to embed the encoded secret message into cover image coefficients. GA is used to find the optimal mapping function for each block. Thus, the hidden capacity and quality of stego image are improved. Finally, the OPAP is applied to reduce the embedding error between the cover image blocks and stego image blocks. This is considered a further improvement of stego image quality. The experimental results are evaluated by four standard images as covers, and with three types of secret messages. The results demonstrate high visual quality of stego image with large size of embedded secret data compared with recent techniques.