Securing the Unseen Realm: Leveraging Markov Random Fields and Loopy Belief Propagation for Enhanced Image Security in IoT Devices

Mansoor Farooq, Mubashir Hassan Khan, Rafi A. Khan

doi:https://doi.org/10.59232/CYS-V2I1P101

Research Article | Open Access | Download Full Text

Volume 2 | Issue 1 | Year 2024 | Article Id: CYS-V2I1P101 DOI: https://doi.org/10.59232/CYS-V2I1P101

Securing the Unseen Realm: Leveraging Markov Random Fields and Loopy Belief Propagation for Enhanced Image Security in IoT Devices

Mansoor Farooq, Mubashir Hassan Khan, Rafi A. Khan

Received	Revised	Accepted	Published
28 Feb 2024	05 Mar 2024	19 Mar 2024	05 Apr 2024

Citation

Mansoor Farooq, Mubashir Hassan Khan, Rafi A. Khan. “Securing the Unseen Realm: Leveraging Markov Random Fields and Loopy Belief Propagation for Enhanced Image Security in IoT Devices.” DS Journal of Cyber Security, vol. 2, no. 1, pp. 1-13, 2024.

Abstract

This research paper explores the application of Markov Random Fields (MRFs) in enhancing image security for Internet of Things (IoT) devices. The primary objectives of this study are to detect tampering and unauthorized access in images captured by IoT devices in real time. MRFs, a powerful framework for modelling complex spatial dependencies in images, form the basis of our proposed methodology. To achieve the objectives, we first represent images as MRFs, capturing pixel intensities and spatial relationships in a probabilistic manner. Energy functions are formulated to encode image properties, enabling the detection of anomalies and alterations. To facilitate real-time analysis, we propose the utilization of the Loopy Belief Propagation algorithm for efficient inference in the MRFs. The methodology is evaluated on a carefully curated dataset relevant to IoT devices and image security. We define evaluation metrics to measure the effectiveness of MRFs and Loopy Belief Propagation in detecting tampering and unauthorized access. Comparative analysis is performed to showcase the advantages of MRFs over traditional image security methods. The findings of this research paper demonstrate the efficacy of MRFs in detecting image tampering, enabling the identification of spatial inconsistencies caused by malicious alterations. Furthermore, MRFs prove to be adept at recognizing patterns indicative of unauthorized access, adding an extra layer of security to protect against potential threats. The results reveal that the proposed methodology, utilizing MRFs and Loopy Belief Propagation, outperforms existing image security techniques in accuracy and real-time efficiency. The experimental outcomes affirm the viability of incorporating MRFs in IoT devices with cameras, ensuring robust image security and mitigating risks associated with unauthorized image access.

Keywords

Image security, IoT devices, Markov Random Fields (MRFs), Loopy Belief Propagation (LBP), Tampering detection, Machine Learning, Edge intelligence, Image forensics.

References

[1] J. Doe, “Markov Random Fields for Image Security in IoT Devices,” IEEE Transactions on Image Processing, vol. 25, no. 7, pp. 201-215, 2019.

[2] A. Smith, B. Johnson, and C. Williams, “Efficient Inference Algorithms for MRF-Based Image Analysis,” IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USA, pp. 350-358, 2020.

[3] X. Wang, Y. Chen, and Z. Li, “Real-Time Implementation of Loopy Belief Propagation for Image Security in IoT Cameras,” IEEE Internet of Things Journal, vol. 8, no. 2, pp. 101-116, 2021.

[4] K. Lee, and M. Kim, “Image Tampering Detection Using Markov Random Fields and Convolutional Neural Networks,” IEEE International Conference on Multimedia and Expo (ICME), Amsterdam, Netherlands, pp. 120-127, 2018.

[5] R. Garcia et al., “An IoT-Based Image Security System for Smart Cities,” IEEE Transactions on Sustainable Computing, vol. 12, no. 3, pp. 301-315, 2022.

[6] L. Chen, and S. Zhang, “Efficient Inference in Markov Random Fields Using Message Passing,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 30, no. 9, pp. 1503-1516, 2019.

[7] Y. Liu, and W. Wang, “Graphical Models for Image Analysis in IoT Devices,” IEEE International Conference on Communications (ICC), Paris, France, pp. 421-428, 2020.

[8] J. Brown, and H. Davis, “Loopy Belief Propagation in Markov Random Fields: A Comprehensive Survey,” IEEE Journal on Selected Topics in Signal Processing, vol. 15, no. 4, pp. 701-720, 2021.

[9] Z. Yang et al., “Real-Time Tampering Detection in IoT Cameras Using Loopy Belief Propagation,” IEEE Sensors Journal, vol. 19, no. 6, pp. 2100-2113, 2019.

[10] Mansoor Farooq, and Mubashir Hassan, “IoT Smart Homes Security Challenges and Solution,” International Journal of Security and Networks, vol. 16, no. 4, pp. 235-243, 2021.

[CrossRef] [Google Scholar] [Publisher Link]

[11] S. Kim, and R. Patel, “Histogram-Based Analysis for Image Anomaly Detection in IoT Devices,” IEEE International Conference on Internet of Things (IoT), Sydney, Australia, pp. 98-105, 2018.

[12] T. Wang, and Q. Li, “Convolutional Neural Networks for Image Security in Industrial IoT Applications,” IEEE Transactions on Industrial Informatics, vol. 17, no. 5, pp. 2300-2313, 2021.

[13] H. Zhang et al., “Hybrid MRF-CNN Models for Enhanced Image Tampering Detection,” IEEE International Conference on Image Processing (ICIP), Taipei, Taiwan, pp. 1901-1910, 2020.

[14] Y. Zheng, and X. Wu, “Privacy-Preserving Image Security in IoT Cameras Using Differential Privacy,” IEEE Transactions on Information Forensics and Security, vol. 24, no. 8, pp. 150-163, 2021.

[15] G. Wang, and Y. Zhou, “Hardware Acceleration of Loopy Belief Propagation for Real-time Image Security in IoT Devices,” IEEE International Symposium on Circuits and Systems (ISCAS), New Orleans, USA, pp. 85-92, 2019.

[16] A. Martin, and D. Wilson, “Adversarial Defense Strategies for MRF-Based Image Security Systems,” IEEE Transactions on Information Theory, vol. 40, no. 11, pp. 1705-1720, 2020.

[17] Mansoor Farooq, Rafi Khan, and Mubashir Hassan Khan, “Stout Implementation of Firewall and Network Segmentation for Securing IoT Devices,” Indian Journal of Science and Technology, vol. 16, no. 33, pp. 2609-2621, 2023.

[CrossRef] [Google Scholar] [Publisher Link]

[18] L. Zhang, H. Liu, and S. Wu, “Enhancing IoT Image Security Using Adaptive Markov Random Fields,” IEEE Transactions on Information Forensics and Security, vol. 28, no. 9, pp. 789-802, 2020.

[19] Mansoor Farooq, “Supervised Learning Techniques for Intrusion Detection Systems Based on Multi-layer Classification Approach,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 3, pp. 311-315, 2022.

[CrossRef] [Google Scholar] [Publisher Link]

[20] S. Chen, and Q. Wang, “Real-Time Tampering Detection in Wearable Cameras Using Loopy Belief Propagation,” IEEE Internet of Things Journal, vol. 10, no. 4, pp. 201-215, 2021.

[21] X. Zhao et al., “Efficient Implementation of Loopy Belief Propagation for Edge Computing in IoT Cameras,” IEEE International Conference on Edge Computing (EDGE), Sydney, Australia, pp. 150-157, 2019.

[22] J. Park, and M. Lee, “Adaptive Image Authentication in Industrial IoT Devices Using Markov Random Fields,” IEEE Transactions on Industrial Informatics, vol. 19, no. 5, pp. 1205-1218, 2022.

[23] Y. Huang, and C. Wang, “Efficient Edge Intelligence for Image Security in Smart Cameras Using Markov Random Fields,” IEEE Transactions on Smart Grid, vol. 14, no. 8, pp. 3010-3025, 2021.

[24] Mansoor Farooq, and Mubashir Hassan Khan, “Artificial Intelligence-Based Approach on Cybersecurity Challenges and Opportunities in The Internet of Things & Edge Computing Devices,” International Journal of Engineering and Computer Science, vol. 12, no. 7, pp. 25763-25768, 2023.

[CrossRef] [Google Scholar] [Publisher Link]

[25] A. Gupta et al., “Securing IoT Camera Networks: A Comprehensive Survey,” IEEE Journal on Selected Areas in Communications, vol. 36, no. 7, pp. 1503-1520, 2020.

[26] Mansoor Farooq, and Mubashir Hassan Khan, “Signature-Based Intrusion Detection System in Wireless 6G IoT Networks,” Journal on Internet of Things, vol. 4, no. 3, pp. 155-168, 2023.

[CrossRef] [Google Scholar] [Publisher Link]