Steganography using an Interlaced Minimax Eigenvector Decomposition Algorithm
Kuan, Shih-Hau (2004) Steganography using an Interlaced Minimax Eigenvector Decomposition Algorithm. Masters thesis, Iowa State University.
Full text available as:
The use of images and audio data for covert channels, copyright protection, and ownership of data has seen an explosion of interest during the past five to ten years. The word steganography is derived from Greek, and it means hiding a secret message in an image or audio file so that the very existence of the message is concealed. Steganography is different from digital watermarking, which is concerned with issues of copyright protection and intellectual property of the image, and it is also different from cryptography, which is the study of scrambling a message so that if and when it is intercepted, it cannot be understood. Steganographic algorithms can be used to create covert communication channels. A covert communication channel is a public medium used to transmit secret information. Images, audio files, and even computer code can be used to transmit secret messages. The thesis presents a steganographic algorithm using a mathematical transform called the Minimax Eigenvector Decomposition (MED). The MED is a nonlinear version of the singular value decomposition. The algorithm is an extension of previous work by Allen given in . Allen’s algorithm had several shortcomings, one of which is that the perceptual change introduced into the stego image was large and visually noticeable. The main contribution of this thesis is that this algorithm overcomes this problem by making the resulting stego image more perceptually close to the original image. It does this by using not only the original image but its transform (viewed as a matrix), creating an interlaced stego image. The result is more visually pleasing. An analysis and discussion of the result applied to ten different images are presented.
Archive Staff Only: edit this record