Introduction to Biometric Systems Ruomu Guo CSPC 620Computer

Introduction to Biometric Systems Ruomu Guo CSPC 620—Computer Security

Overview Identification Person’s body && identity Applications: Fingerprint, Iris, Retina Recognition, Face Detection, Hand Geometry. Relationship between Biometric System and certain topics in the area of computer security.

Overview Refers to the use of mathematical statistical methods to analyze biological behaviors or characteristics. Biometric System mainly consists of four modules— Sensor, Feature Extraction, Matcher, System Database Storage. Difficulties: Accuracy, Speed, Resource Requirements, Harmless to human beings, Robust to fraudulent attacks.

Measurement of Biometric System Fig: The relationship between FAR, FRR, and Threshold Value

Fingerprint Recognition, because of its lifetime invariance, uniqueness and convenience, is becoming an important method for biometric identification. Finger skin ridge and valley forms a regular array of different pattern types. Valley, ridge combined with point and bifurcation point, are called the fingerprint minutiae points (minutiae). By comparing different fingerprint minutiae, person’s identity can be recognized or identified.

Fingerprint Recognition Fig: Block Diagram of Fingerprint Recognition Processes

Fingerprint Recognition Sense: off-line fingerprint acquisition, live-scan sensing. Feature Extraction: singular region, local ridge orientation Matching: Correlation-based matching, Minutiae-based matching, Ridge feature-based matching Database Storage: update periodically

Face Detection is also a popular method of biometric system for recognition and identifies individual’s identity. Advantages: widely accept as a identifier, least intrusive. Disadvantages: illumination, disguise for circumvention, and incompatible with pure identification protocol.

Face Detection Primary methods for detecting faces 1. Knowledge-based methods 2. Feature invariant approaches 3. Template matching methods 4. Appearance-based methods

Face Detection The technique for face recognition can be classified as following three groups: 1. Feature Methods: 2. Holistic Methods: 3. Hybrid Methods:

PCA Application PCA (Principal Component Analysis) A face image usually defines a point in the highdimensional image space. PCA is used to simplify the required process of analysis by reducing the dimensional spaces or subspaces.

Conclusion Biometric System is not independent as a module for entire computer security area. Some ticklish problems in computer security will be solved appropriately such as authentication for each person’s identity before they will enter or access to other systems. Scientists are still trying to exploit other methods to improve the performance of biometric system with more enhancement of computer security.

Reference 1. A. K. Jain, A. Ross and S. Prabhakar, An Introduction to Biometric System IEEE Transactions on Circuits and Systems for Video Technology, Special Issue on Image- and Video-Based Biometrics, Vol. 14, No. 1, pp. 4 -20, January 2004. 2. A. Jain, R. Bolle, and S. Pankanti, Introduction to Biometrics: Personal Identification in Networked Society (A. Jain, R. Bolle, and S. Pankanti, Eds. ), pp. 1 -41, Boston, MA: Kluwer Academic, 1999. 3. D. Maltoni. A tutorial on Fingerprint Recognition: In M. Tistarelli, J. Bigun, and E. Grosso, editors, Biometrics School 2003, LNCS 3161, pages 43 -68. Springer Verlag, Berlin, Heidelberg, 2005. 4. Description of Face Detection at Wikipedia: http: //en. wikipedia. org/wiki/Face_detection.