Image steganography is a common form of information hiding which embeds as many message bits into images and keep the introduced distortion imperceptible. How to balance the trade-off between the capacity and imperceptibility has become a very important issue in the researches of steganography. In this chapter, we discuss one kind of the solution for disposing the trade-off, named adaptive image steganography. After a brief review, we present two methods based on structural similarity metric. The first one is based on the generalized LSB, in which the substitution depth vector is obtained via the dynamic programming under the constraint of an allowable distortion. The second method is proposed to use adaptive quantization-embedder to carry message bits. Different from the first method, the distortion index is constructed by contrast-correlation distortion. The other difference is that the parameters of the adaptive quantization embedder are embedded into the image containing message bits by the reversible da a hiding method. Beside that, we also bring forward some attractive directions worthy of being studied in the future. Furthermore, we find that the existing methods do not have a good way to control the amount of information and the distortion as an extract manner, and most schemes are designed just according to the experiences and experiments.
Steganography is the art of covered or hidden writing. The practice of steganography has quite a history. The earliest example can be dated to the Greek historian Herodotus, who describes how one of his cunning countrymen sent a secret message warning of an invasion by scrawling it on the wood underneath a wax table. To casual observers, the tablet appeared blank. Roman empires would hide secret message in ordinary message that only a certain formula when applied could reveal the real message. This was of course to prevent the real message from falling into the wrong hands. Both Axis and Allied spies during World War II used such measure as invisible inks as tiny punctures above key characters in a document that form a message when combined or ‘invisible inks ’ that could only be read when heat was applied.
The earliest study concerning modern steganography was present by G. J. Simmons (Simmons, 1983). In (Simmons, 1983), the famous story of Prisoners’ problem explains what capabilities and merits steganography has to offer when public communication channel is insecure. The story is about two prisoners Alice and Bob who are put to different cells in a jail and what to plan to escape form the prison together as Figure 1 illustrates. However, any communications between them have to pass though the warden Eve, so only plaintext (i.e. content that have clear meanings) can be handed on by Eve, while anything in the form of ciphertext(i.e. contents that seems meaningless) will not be allowed. Bob and Alice want to conspire to escape, but they can not send a clear message to each other showing the intention of escaping, and neither can they encrypt the secret message about their plan of escape into a cipher text in an attempt to hide their intention before the message comes to Eve. To solve this problem, Alice can use a picture (like illustrated in Figure 1) as the cover to conceal the very existence of the secret message from Eve.
It is worthy to be noticed that often dictionaries have defined steganography the same as cryptography, however there is a clear difference between the two. The purpose of steganography is covert communication to hide the existence of a message from a third party, which differs from cryptography, the art of secret writing, which is intended to make a message unreadable by a third party but does not hide the existence of the secret communication. The steganographic process generally involves placing a hidden message in some transport medium, called the carrier or the cover medium. The secret message is embedded in the cover medium to form the stego medium. The use of a stego key may be employed for encryption of the hidden message or for randomization of the embedding locations in the steganographic scheme. The whole process to embedding message bits into images can be depicted by Figure 2.
Nowadays, digital steganography has been a specialized field to research the art and science of how to conceal a secret message in various kinds of media, such as image, audio, video and so on. Steganography essentially is a multidisciplinary field combining use of image and signal processing techniques with cryptography, communication theory, coding theory, signal compression theory of visual perception, etc(Fridrich, 1999).
Key Terms in this Chapter
Adaptive Steganography: In the chapter, the adaptive steganography is defined as the mechanism to choose the amount of the bits for hiding according to the HVS characteristics of the target pixel of a group of pixels.
Structural Similarity Metric: The structural similarity metric is the image quality index proposed by Wang and Bovik at about 2000, which models any distortion as a combination of the loss of correlation, mean distortion and variance distortion. The Q index and SSIM are two realization of the structural similarity metric.
Steganography: The technique to hide the secret message into the cover object to realized the covert communication without being found by the third part, which can conceal the existence of the communication itself.
SDV (Substitution Depth Vector): The SDV is defined in the part of generalized-LSB-based adaptive steganography in the chapter, which denotes the substitution depths assigned for all image blocks. Under the given allowance distortion, SDV can be calculated according to the proposed algorithm in the chapter.
AQE (Adaptive Quantization Embedder): The AQE is the embedding function that can be used to make adaptive steganography. With the changes of the sizes of the quantizers of AQE, the amount of message bits that can be embedded changes.
Contrast-Correlation Distortion: The constrast-correlation distortion is defined as the combination of the constrast distortion and correlation distortion, which are both borrowed from the structural similarity metrics.
Imperceptibility: The imperceptibility is defined as the perceptual quality distortion between the cover media and the stego one. For image, the PSNR, wPSNR, Waston Metric and SSIM are usually used to depict the distortion cause by the steganography.
Capactity: The capacity denotes the amount of the message bits that can be embedded into the cover object under the given constraints such as imperceptibility, security and robustness etc.
Complete Chapter List
Shiguo Lian, Yan Zhang
Shiguo Lian, Yan Zhang
Pramod A. Jamkhedkar, Gregory L. Heileman
Deepali Brahmbhatt, Mark Stamp
Mercè Serra Joan, Bert Greevenbosch, Anja Becker, Harald Fuchs
Hugo Jonker, Sjouke Mauw
Pallavi Priyadarshini, Mark Stamp
L. Badia, A. Erta, U. Malesci
Ramya Venkataramu, Mark Stamp
Nicolas Anciaux, Luc Bouganim, Philippe Pucheral
Guojun Wang, Yirong Wu, Geyong Min, Ronghua Shi
Supavadee Aramvith, Rhandley D. Cajote
M. Hassan Shirali-Shahreza, Mohammad Shirali-Shahreza
Pradeep K. Atrey, Abdulmotaleb El Saddik, Mohan Kankanhalli
Esther Palomar, Juan M.E. Tapiador, Julio C. Hernandez-Castro, Arturo Ribagorda
Andreas U. Schmidt, Nicolai Kuntze
Goo-Rak Kwon, Sung-Jea Ko
Frank Y. Shih, Yi-Ta Wu
Guangjie Liu, Shiguo Lian, Yuewei Dai, Zhiquan Wang
Minglei Liu, Ce Zhu
Hsuan T. Chang, Chih-Chung Hsu