SUDS: Sanitizing Universal and Dependent Steganography
This addresses the security challenge of protecting against covert communication by bad actors, offering a novel approach that is not dependent on known steganographic signatures, though it appears incremental as it builds on deep learning techniques for sanitization.
The paper tackles the problem of detecting and sanitizing steganography (hidden messages) in digital media, which is increasingly used by malicious actors, by proposing SUDS, a deep learning sanitization technique that does not rely on prior knowledge of hiding methods. It demonstrates SUDS's effectiveness by testing it on various steganographic methods and showing it can increase resistance against attacks by 1375% in a real-world scenario.
Steganography, or hiding messages in plain sight, is a form of information hiding that is most commonly used for covert communication. As modern steganographic mediums include images, text, audio, and video, this communication method is being increasingly used by bad actors to propagate malware, exfiltrate data, and discreetly communicate. Current protection mechanisms rely upon steganalysis, or the detection of steganography, but these approaches are dependent upon prior knowledge, such as steganographic signatures from publicly available tools and statistical knowledge about known hiding methods. These dependencies render steganalysis useless against new or unique hiding methods, which are becoming increasingly common with the application of deep learning models. To mitigate the shortcomings of steganalysis, this work focuses on a deep learning sanitization technique called SUDS that is not reliant upon knowledge of steganographic hiding techniques and is able to sanitize universal and dependent steganography. SUDS is tested using least significant bit method (LSB), dependent deep hiding (DDH), and universal deep hiding (UDH). We demonstrate the capabilities and limitations of SUDS by answering five research questions, including baseline comparisons and an ablation study. Additionally, we apply SUDS to a real-world scenario, where it is able to increase the resistance of a poisoned classifier against attacks by 1375%.