Designing a robust image steganography algorithm for secure cyberinfrastructure

There is an enormous need to overcome the increasing challenges of transferring sensitive multimedia data including images over cloud infrastructure platforms in a robust but secure way. Previous solutions to this challenge have been to encrypt sensitive image data before transmission and decrypt upon retrieval of the data. Although encrypted data may not be interpreted by unlawful information hunters, this method has been proved to be unsafe because it is difficult to conceal. In many cases, encryption may not be permitted or appropriate. Furthermore, it is nearly impossible to develop and implement an encryption-decryption scheme for a secure cyberinfrastructure system that will not be compromised over time. Recently, a newer approach to conceal sensitive data called digital steganography has been developed to augment the failures of encryption schemes. Unlike encryption which does not hide the presence of sensitive data but makes the data inaccessible, steganography is intended to conceal the existence of such data. Hence steganography is very useful in secure cyber-systems. This thesis aims at developing a new and simple steganography algorithm to embed pictures known as payload images, into image files known as cover images to get stego images. The algorithm used is a novel combination of some of the previous steganography techniques known as the Discrete Cosine Transform (DCT) and the Least Significant Bit (LSB). The results of the novel method used in this thesis was found to be more effective than some previous methods using the Signal to Noise Ratio (SNR) as the basis of comparison