Evaluation of pixel-based steganography and stegodetection methods

摘要

Steganography is a process that involves hiding a message in an appropriate carrier, for example an image or an audio file. The carrier can then be sent to a receiver without anyone else knowing that it contains a hidden message. This is a process which can be used, for example, by civil rights organizations in repressive states to communicate their message to the outside world without their own government's being aware of it. Less virtuously, it can be used by terrorists to communicate with one another without other people's knowledge. In both cases, the objective is not to make it difficult to read the message as cryptography does, but to hide the existence of the message in the first place, possibly to protect the courier. The aim of this study was to investigate, implement and evaluate a number of common methods of pixel-based steganography. The strengths and weaknesses of the chosen methods were analysed. To provide a common frame of reference, all the steganography methods implemented and analysed worked with GIF images. Eight steganography methods were implemented. The methods were chosen for their different strengths in terms of resistance to different types of steganalysis or their ability to maximize the size of the message they could store. All the methods used were based on the manipulation of the least significant bits of pixel values or the rearrangement of colours to create least significant bit(s) or parity patterns, which correspond to the message being hidden.