LIST OF FIGURES iv LIST OF TABLES

Steganography is a science of hiding messages into multimedia documents. A message can be hidden in a document only if the content of a document has high redundancy. Although the embedded message changes the characteristics and nature of the document, it is required that these changes are difficult to be identified by an unsuspecting user. On the other hand, steganalysis develops theories, methods and techniques that can be used to detect hidden messages in multimedia documents. The documents without any hidden messages are called cover documents and the documents with hidden messages are named stego documents. The work of this thesis concentrates on image steganalysis. We present four different types of steganalysis techniques. These steganalysis techniques are developed to counteract the steganographic methods that use binary (black and white) images as the cover media. Unlike greyscale and colour images, binary images have a rather modest statistical nature. This makes it difficult to apply directly the existing steganalysis on binary images. The first steganalysis technique addresses blind steganalysis. Its objective is to detect the existence of a secret message in a binary image. Since the detection of a secret message is often modelled as a classification problem, consequently it can be approached using pattern recognition methodology. The second steganalysis technique is known as multi-class steganalysis. Its purpose is to identify the type of steganographic method used to create the stego image. This extends the earlier blind steganalysis from two-class (cover or stego image) to multi-class (cover or different types of stego images) classification. Similar to blind steganalysis, this technique is also based on the pattern recognition methodology to perform the classification. The third steganalysis technique uses first-order statistic—binary pattern histogram—to estimate the length of an embedded message. This technique is used specifically to analyse the steganography developed by Liang et al. The estimated message length usually plays an important role and is needed at other levels of analysis. The fourth steganalysis technique identifies the steganographic payload locations based on multiple stego images. This technique can reveal which pixels in the binary image carry the message bits. This technique is crucial as it not only