Targeted Detection of Steganographic Content in Images Using Transfer Learning to Enhance Cybersecurity

Revannaradhya, Nisarga (2024) Targeted Detection of Steganographic Content in Images Using Transfer Learning to Enhance Cybersecurity. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (2MB) | Preview

Abstract

Steganography is a technique used to conceal information within digital media while making it unrecognizable. It can serve for both secure communication and malicious purposes, such as embedding malware in images for cyberattacks. Through image steganography, malicious code can be hidden in natural images without arousing suspicion. Steganalysis is the counter process of detecting steganographic content. This work introduces an image steganalysis approach designed to identify images embedded with JavaScript code using spread spectrum steganography. This is done by combining traditional feature extraction methods with pre-trained models, enhanced by an attention mechanism. However, despite the advancements, CNNs still tend to have some difficulty in recognizing subtle changes introduced by spread-spectrum steganography, particularly when handling JavaScript code embedding. This limitation highlights the value of the combined approach implemented in this research. Among others, EfficientNet-b0 model achieved 94.7% accuracy when trained on ImageNet, thus proving the effectiveness of the combined approach. This approach can serve as a benchmark to detect the steganographic data hidden within images.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Jameel Syed, Muslim UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	25 Aug 2025 10:40
Last Modified:	25 Aug 2025 10:40
URI:	https://norma.ncirl.ie/id/eprint/8619

Actions (login required)

View Item