APSIPA Transactions on Signal and Information Processing > Vol 12 > Issue 1

Exploring Human Biometrics: A Focus on Security Concerns and Deep Neural Networks

Waleed H. Abdulla, The University of Auckand, New Zealand, w.abdulla@auckland.ac.nz , Felix Marattukalam, The University of Auckand, New Zealand, Vedrana Krivokuća Hahn, Idiap Research Institute, Switzerland
Suggested Citation
Waleed H. Abdulla, Felix Marattukalam and Vedrana Krivokuća Hahn (2023), "Exploring Human Biometrics: A Focus on Security Concerns and Deep Neural Networks", APSIPA Transactions on Signal and Information Processing: Vol. 12: No. 1, e38. http://dx.doi.org/10.1561/116.00000021

Publication Date: 06 Sep 2023
© 2023 W. H. Abdulla, F. Marattukalam and V. Krivokuća Hahn
Biometricscomputer visiondeep learningtemplate protection and recognition


Open Access

This is published under the terms of CC BY-NC.

Downloaded: 804 times

In this article:
Human Biometrics Definitions 
Biometrics Recognition Systems Operative Modes OR Biometric Features and Types of Biometrics 
Biometrics versus Classical Recognition 
Biometrics Favored Attributes 
Applications of Biometrics 
Common Questions on the Design of Biometric Systems 
Biometric Technology 
Structures of the Human Biometric Recognition Systems 
Positive and Negative Recognition 
Biometric System Performance Measures 
Biometrics Fusion 
Multibiometric System Design Considerations 
Biometric Template Protection 
Deep Learning Approach to Developing Human Biometric Systems 
Conclusion and Future Perspective 


Biometric technology is rapidly growing due to the urgent need to secure people’s properties, from goods to information, in the overwhelming digital technology proliferation in all aspects of society. In this paper, biometric recognition is defined as the automated recognition of individuals based on biological or behavioral characteristics, such as fingerprints, facial recognition, and speech patterns. The authors emphasize that a robust biometric system consists of a combination of physiological and behavioral features. However, using biometrics for identification raises privacy concerns and the paper addresses the need to balance privacy and security. A comprehensive section on biometric template protection is introduced to address biometrics privacy and different attack protections. It discusses deep neural network-based models to segment real-world features and match them for authentication. It presents a case study of a new model based on the Siamese neural network. It explains how the Siamese neural network can be used for biometric recognition and how it compares to other deep learning models commonly used in the field. Lastly, the paper discusses state-of-the-art methods to secure information and provides a futuristic view of the technology. This paper provides a comprehensive overview of biometric technology, its advantages, and the associated privacy concerns.