Artificial Intelligence for Biometrics and Cybersecurity: Technology and applications

✍ Scribed by Ahmed A. Abd El-Latif, Mohammed Adel Hammad, Yassine Maleh, Brij B. Gupta, Wojciech Mazurczyk

• Publisher: The Institution of Engineering and Technology
• Year: 2023
• Tongue: English
• Leaves: 290
• Series: the
• Category: Library

✦ Synopsis

he integration of new technologies is resulting in an increased demand for security and authentication in all types of data communications. Cybersecurity is the protection of networks and systems from theft. Biometric technologies use unique traits of particular parts of the body such facial recognition, iris, fingerprints and voice to identify individuals' physical and behavioural characteristics. Although there are many challenges associated with extracting, storing and processing such data, biometric and cybersecurity technologies along with artificial intelligence (AI) are offering new approaches to verification procedures and mitigating security risks.

This book presents cutting-edge research on the use of AI for biometrics and cybersecurity including machine and deep learning architectures, emerging applications and ethical and legal concerns. Topics include federated learning for enhanced cybersecurity; artificial intelligence-based biometric authentication using ECG signal; deep learning for email phishing detection methods; biometrics for secured IoT systems; intelligent authentication using graphical one-time-passwords; and AI in social cybersecurity.

Artificial Intelligence for Biometrics and Cybersecurity: Technology and applications is aimed at artificial intelligence, biometrics and cybersecurity experts, industry and academic researchers, network security engineers, cybersecurity professionals, and advanced students and newcomers to the field interested in the newest advancements in artificial intelligence for cybersecurity and biometrics.

✦ Table of Contents

Cover
Contents
About the editors
Preface
1 Introduction
References
2 AI in biometrics and cybersecurity
2.1 Introduction
2.2 History of biometric systems
2.3 Challenges of biometric systems
2.4 Terminologies and fundamental concepts of biometric systems
2.5 AI in biometric security systems and cybersecurity
2.6 Conclusion
References
3 Biometric security performance: analysis methods and tools for evaluation and assessment
3.1 Introduction
3.2 Background
3.3 Performance evaluation metrics
3.4 Biometric performance evaluation methods
3.4.1 Test and evaluation standards
3.4.2 Verification and identification performance metrics
3.4.3 Performance testing procedures
3.4.4 Performance testing protocols
3.5 Biometric system performance analysis
3.5.1 Performance metrics for biometric systems
3.5.2 Performance analysis techniques
3.5.3 Accuracy, speed, and robustness testing
3.6 Biometric security evaluation
3.6.1 Biometric system vulnerabilities
3.6.2 Threat models and attacks
3.6.3 Countermeasures and mitigation techniques
3.7 Biometric security performance evaluation tools
3.7.1 Open-source and proprietary tools
3.7.2 Tools for performance testing, analysis, and evaluation
3.7.3 Tools for security evaluation and assessment
3.8 Case studies
3.8.1 Case studies of biometric security performance evaluation
3.8.2 Best practices in biometric security performance evaluation
3.8.3 Lessons learned from biometric security performance evaluation
3.9 Conclusion
Appendix A
Appendix B
Appendix C
References
4 Leveraging generative adversarial networks and federated learning for enhanced cybersecurity: a concise review
4.1 Introduction
4.2 Mathematical formulation
4.3 GAN and federated learning for image security
4.3.1 GAN for image security
4.3.2 Federated learning for image security
4.3.3 Combining GANs with FL
4.4 GAN and federated learning for cybersecurity
4.5 Evaluation and performance analysis
4.6 Open challenges in leveraging GANs and FL for cybersecurity
4.7 Conclusion and future directions
Appendix A
References
5 A survey on face recognition methods with federated leaning
5.1 Introduction
5.2 Federated learning
5.2.1 Concept and definition of federated learning
5.2.2 Classification of federated learning
5.3 Face recognition
5.3.1 The concept of face recognition
5.3.2 Face recognition database
5.3.3 Face recognition technology process
5.3.4 Advantages and limitations of different face recognition approaches
5.4 The current development of FedFR technology
5.4.1 FedFR with privacy protection
5.4.2 Personalized FedFR
5.5 Hybrid approaches and ensemble methods
5.5.1 Introduction to hybrid approaches in face recognition
5.5.2 Discussion of ensemble methods in face recognition
5.6 Conclusion
5.7 Future outlook
Appendix A
Appendix B
References
6 Artificial intelligence-based biometric authentication using ECG signal
6.1 Introduction
6.1.1 Overview of biometrics
6.2 Related works on AI-based ECG biometrics
6.2.1 ECG biometrics based on fiducial features
6.2.2 ECG biometrics based on non-fiducial features
6.2.3 ECG biometrics based on hybrid features
6.2.4 ECG biometrics based on other approaches
6.3 Materials and methods
6.3.1 ECG database
6.3.2 ECG data pre-processing
6.3.3 ECG fiducial points detection
6.3.4 ECG feature extraction
6.3.5 ECG feature optimization or selection (optional)
6.3.6 ECG signal classification and identification
6.4 Experimental results
6.4.1 Performance metrics of the classifiers
6.4.2 Performance parameters for classical model classifiers
6.4.3 Performance parameters for graphical model classifiers
6.4.4 Performance parameters for ensemble model classifiers
6.4.5 Overall comparison of AI-based machine learning classifiers for ECG biometrics
6.5 Conclusion
References
7 A comparative analysis of email phishing detection methods: a deep learning perspective
7.1 Introduction
7.2 Related work
7.2.1 Traditional phishing detection methods
7.2.2 Deep learning approaches for phishing detection
7.3 Data collection and preprocessing
7.4 Methodology
7.4.1 Word embedding
7.4.2 Evaluation criteria
7.5 Results and analysis
7.5.1 With CNN model
7.5.2 With RNN model
7.5.3 With LSTM model
7.5.4 With Bi-LSTM model
7.6 Discussion
7.7 Conclusion
References
8 Securing hardware coprocessors against piracy using biometrics for secured IoT systems
8.1 Introduction
8.2 IoT hardware security, threat model, and its significance
8.3 Fingerprint biometric-based hardware security
8.3.1 Overview of the fingerprint biometric-based approach
8.3.2 Details of the fingerprint biometric-based methodology
8.3.3 Advantages of the fingerprint (signature-based) biometric approach
8.4 Facial biometric-based hardware security
8.4.1 Detection of counterfeited hardware using the facial biometric approach
8.5 Contactless palmprint biometric-based hardware security
8.5.1 Overview of the contactless palmprint biometric approach
8.5.2 Details of the contactless palmprint biometric methodology [11]
8.5.3 Advantages of contactless palmprint biometric security [11] technique over various others
8.6 Other security techniques used for IoT security and identity authentication
8.7 Comparative perspective and analysis
8.8 Conclusion
Acknowledgement
References
9 Intelligent authentication system using graphical one-time passwords
9.1 Introduction
9.2 Alphanumeric password-based authentication
9.3 Alternative authentication systems
9.3.1 One-time passwords
9.3.2 Certificate-based authentication
9.3.3 Biometric authentication
9.3.4 Graphical password authentication
9.4 Graphical passwords
9.4.1 Recognition-based GPS
9.5 Graphical OTP system
9.5.1 Registration phase
9.5.2 Authentication phase
9.6 Measurements
9.7 Analysis and discussion of results
9.8 Conclusion
9.9 Future work
References
10 Role of AI in social cybersecurity: real-world case studies
10.1 Introduction
10.2 The rise of social media and its implications
10.2.1 Social media and cybersecurity challenges
10.2.2 Implications of social cybersecurity
10.2.3 The necessity of implementing social cybersecurity measures
10.3 Current scenario of AI in social cybersecurity
10.4 AI fake profile and bot detection
10.4.1 The rise of fake profiles and bots
10.4.2 AI-based fake profile detection
10.4.3 Challenges and limitations
10.5 AI in sexual predator detection
10.6 Case study on Facebook and Twitter
10.6.1 Case study 1: Twitter’s AI-powered fake profile identification
10.6.2 Case study 2: Facebook’s AI-powered fake profile detection
10.7 Conclusion
Appendix A
References
11 Ethical and privacy concerns and challenges
11.1 Introduction
11.2 Ethical considerations
11.3 Privacy concerns
11.4 Responsible AI practices
11.5 Mitigation strategies
11.6 Collaboration and governance
11.7 Future directions and recommendations
11.8 Conclusion
References
12 Conclusion
Index