A Statistical Evaluation Model for Minutiae-Based Automatic Fingerprint Verification Systems
Evaluation of the reliability of an Automatic Fingerprint Verification System (AFVS) is usually performed by applying it to a fingerprint database to get the verification accuracy. However, such an evaluation process might be quite time consuming especially for big fingerprint databases. This may prolong the developing cycles of AFVSs and thus increase the cost. Also, comparison of the reliability of different AFVSs may be unfair if different fingerprint databases are used. In this paper, we propose a solution to solve these problems by creating an AFVS evaluation model which can be used for verification accuracy prediction and fair reliability comparison. Experimental results show that our model can predict the performance of a real AFVS pretty satisfactorily.
KeywordsProbability Density Function Fingerprint Image Direction Difference Core Point Verification Performance
- 2.Galton, F.: Finger Prints. MacMillan, Basingstoke (1892)Google Scholar
- 3.Trauring, M.: Automatic Comparison of Finger Ridge Patterns. Nature, 938–940 (1963)Google Scholar
- 4.Stoney, D.A., Thornton, J.I.: A Critical Analysis of Quantitative Fingerprint Individuality Models. J. Forensic Sciences 31(4), 1187–1216 (1986)Google Scholar