Biometric Authentication System

Project: $DO NOT EDIT THIS FIELD$  
Supervisor: Ke CHEN
Difficulty grading: SH/INF=U, BM=N, CM=C
Area:    Artificial Intelligence
Max number of students who can do this project: 2

Due to the tragedy of September-11-2001 in New York City and July-7-2005 in London, automatic biometric authentication systems are highly demanded for security. Therefore, the development of such systems of high performance is of great significance and huge potential applications.Unlike traditional authentication techniques, e.g. PIN code or password, biometrics provides an alternative yet natural way for personal identity authentication. Biometrics handles authentication of individuals on the basis of biological and/or behavioral characteristics (measurements of the human body). As a primary advantage, biometric features are typically unique and, therefore, cannot be misplaced and forgotten since these are always inherently associated with human beings. The major biometric features include voice, face, fingerprints, irises, retinas, palmprints, signature, gait and so on.
In general, there are two types of biometric systems: identification for identifying an unknown biometric token as belonging to one of people registered in the system (e.g., a senario of police investigation) and verification for accepting or rejecting the identity claim of a person based on an input biometric token (e.g., a senario of identity authentication during banking). In this project, building an operational prototype based on one of aforementioned biometric features is thoroughly studied. In the process of development, there are numerous engineering issues, e.g. trade-off between performance and acceptability, and robustness on miscellaneous mismatch conditions, to be investigated.

REFERENCES:

  1. A. K. Jain, A. Ross and S. Pankanti, "Biometrics: A tool for information security," IEEE Transactions on Information Forensics and Security, vol. 1, no. 2, 2006, pp. 125-143. (comments: a good survey for biometic systems)
  2. A. K. Jain, R. P. W. Duin and J. Mao, "Statistical pattern recognition: A review," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 1, 2000, pp. 4-37. (comments: a good survey for underpining techniques of biometric systems)

COURSE PREREQUISITES: Machine Learning (Pattern Recognition), Image Processing or Speech Signal Processing (depending on what biometric feature is going to be used in a biometric system)

EQUIPMENT: PC, appropriate I/O device, Matlab (ideal with appropriate toolkits)