Dual model for biometric identification system

Abstract

A biometric system which relies only on a single biometric identifier in making a personal identification is often not able to meet the desired performance requirements. Dual modal has great demands to overcome the issue involved in single biometric identifier. The use of personal identity verification systems with a serial architecture dual-modal biometrics has been proposed using fingerprint and iris pattern in order to increase the performance and security against environmental variations and fraudulent. The dual modal system is based on an empirical analysis of the fingerprint and iris images and it is split in several steps using local image properties. The fingerprint Minutiae extraction steps are loading the finger image, enhancement by histogram equalization and Fourier transform, binarization, segmentation by block direction estimation and region of interest extraction by morphological operations, remove H-break, remove spur, extract minutia and remove false minutia. At the same time, the iris system steps are capturing iris patterns, determine the location of the iris boundaries, converting the iris boundary to the stretched polar coordinate system, extracting the iris code. The dual modal system is achieved at the fingerprint matching score level less than 75%. The implementation of methods, algorithms and the graphical user interface was done by using MATLAB and CASIA database of 11 samples of fingerprint and iris data. Statistical Experimental used in this project on a small sample size, which is very difficult to conduct a full analysis of the observed results and consider as a main limitation of dual modal system. Future work can be done by performing the statistical experiments on a larger sample size, and conduct a full analysis of the observed results.