SUDS: Sanitizing Universal and Dependent Steganography

TL;DR

This paper introduces SUDS, a deep learning-based sanitization method that effectively detects and neutralizes various steganography techniques without prior knowledge, enhancing security against covert data exfiltration and malware communication.

Contribution

SUDS is a novel deep learning approach that sanitizes both universal and dependent steganography, overcoming limitations of traditional steganalysis reliant on prior signatures.

Findings

SUDS effectively detects and sanitizes LSB, DDH, and UDH steganography methods.

SUDS increases resistance of a poisoned classifier against attacks by 1375%.

The paper provides comprehensive evaluations including baseline comparisons and ablation studies.

Abstract

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%.