An effective system to verify human identity is a major challenge in most traditional access control systems in developing countries, as it can easily be compromised. From the vulnerability of banking transactions to the multiple registration in most civic identification projects like voters' registration, Payroll System and Pension Scheme underscores the urgent need for an effective system that provide immediate technological solution. This research work presents an Automated Fingerprint Verification System simulating both Minutiae Based Matching and Cross Correlation Coefficient Matching that provides an effective and efficient means of verifying human identity which significantly decreases the possibility of fraud in access control. The method used MATLAB simulation to align the minutiae of the two-fingerprint image (query template) and stored templates (reference template) inputted to find the total number of minutiae matched. After alignment, two minutiae are considered for matching when spatial distance and direction difference between them are not up to a given tolerance. Finally, the templates were further verified with cross correlation algorithm to improve the result accuracy. This approach has better performance as compared to individual matching technique. The result obtained after testing several fingerprints for identification proved to be efficient by verifying correctly the identities of the persons enrolled and achieving a matching score above 80% threshold for Matching Pair and below 80% threshold for Non-Matching Pair.
展开▼
