The application of DL techniques in steganography has greatly increased the efficiency of steganographic concealment systems, leading to a substantial improvement in safety risks. To counter this, the researchers shifted to steganalysis, the method of detecting steganographic material. In practice, steganalysis is being used in a security framework that operates like a spam-filter to prevent the transmission of malicious steganographic content through networks. Newer steganalysis methods, however, have shown promising results, but they remain both fairly inefficient and non-robust.
Steganography is the mechanism by which data are hidden inside an ordinary (non-secret) file to evade detection. Although encryption is intended to hide data information, steganography is intended to conceal data existence. Steganography can also hide new form of communication by hiding the existence of data, and thus provide behavioral protection. Steganography provides behavioral protection and plays a critical role in safeguarding the privacy of information.
Although steganographic concealment systems are routinely used for benign activities, bad actors can also use them to transmit malicious information using ordinary files such as photographs, thus posing a risk to security. Because current network defenses do not search images for steganographic content, the transmission of malicious steganographic content cannot be effectively blocked. Hackers may thus use steganography to transfer vulnerabilities or other compromising data in an untraceable fashion through networks.
Inhaltsverzeichnis (Table of Contents)
- INTRODUCTION.
- Steganography.
- Steganalysis.
- Deep Learning Overview.
- Contributions.
- Thesis Roadmap.
- LITERATURE SURVEY.
- Introduction.
- Related Literature.
- Analysis of image steganalysis frameworks.
- Summary.
- PROBLEM FORMULATION AND RESEARCH METHODOLOGY.
- Introduction.
- Research Challenges.
- Broad problem statement.
- Research Objectives.
- Research Methodology.
- Summary.
- PROPOSED IMAGE STEGANALYSIS SCHEME.
- Introduction.
- The CNN based framework for image steganalysis.
- Training and testing of proposed image steganalysis framework.
- Summary.
- EXPERIMENTAL ANALYSIS AND RESULTS.
- Introduction.
- Datasets.
- Performance metrics used.
- Summary of the model.
- Experimental Results.
- Comparative Analysis.
- Summary.
- CONCLUSION AND FUTURE WORK.
- Introduction.
- Conclusion and future direction of the research work.
- REFERENCES.
Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)
This thesis explores the use of deep learning techniques to enhance image steganalysis, a process aimed at detecting hidden information embedded within digital images. The primary goal is to develop a robust and efficient steganalysis framework capable of effectively identifying steganographic content, particularly in the context of deep learning-based steganography methods.
- The Evolution of Steganography and Steganalysis
- Deep Learning Approaches to Image Steganalysis
- Developing a Robust Steganalysis Framework
- Performance Evaluation and Comparison
- Future Directions and Applications
Zusammenfassung der Kapitel (Chapter Summaries)
- Chapter 1: Introduction - This chapter provides an overview of steganography, the art of hiding data within seemingly ordinary files, and its counterpart, steganalysis, the process of detecting such hidden information. The chapter also highlights the increasing relevance of deep learning in both steganography and steganalysis, outlining the challenges and opportunities presented by these advanced techniques.
- Chapter 2: Literature Survey - This chapter delves into the existing research literature related to image steganalysis, focusing on the development and evaluation of various steganalysis frameworks. It presents a critical analysis of the state-of-the-art techniques, their limitations, and future directions.
- Chapter 3: Problem Formulation and Research Methodology - This chapter lays out the specific research challenges addressed in this thesis and defines the broad problem statement guiding the research. It outlines the objectives of the research and presents the methodology employed to develop and evaluate the proposed steganalysis framework.
- Chapter 4: Proposed Image Steganalysis Scheme - This chapter details the novel image steganalysis scheme proposed in this thesis. It describes the architecture of the Convolutional Neural Network (CNN)-based framework and explains the principles behind its design. The chapter also elaborates on the training and testing procedures employed to evaluate the framework's performance.
- Chapter 5: Experimental Analysis and Results - This chapter presents the experimental results obtained by applying the proposed steganalysis framework to various image datasets. It details the performance metrics used to evaluate the framework's effectiveness and discusses the comparative analysis against existing methods.
Schlüsselwörter (Keywords)
This work focuses on the application of deep learning techniques to the domain of image steganalysis, examining the challenges and opportunities presented by these advancements. Key terms include image steganalysis, convolutional neural networks (CNNs), steganographic embedding, deep learning, stego-detection, and robustness evaluation.
- Quote paper
- Numrena Farooq (Author), Adil Bashir (Author), Arvind Selwal (Author), 2020, Development of Intelligent Framework. Image Steganalysis using Deep Level Features, Munich, GRIN Verlag, https://www.grin.com/document/1254950
