| Applications of fingerprint recognition as means of identity authentication are becoming more commonly acceptable in several areas, such as financial transactions and identity check at entry borders, to mention a few. Issues related to accuracy, security and processing time impose major challenges to the fingerprint recognition systems.;To address the above challenges in the traditional fingerprint recognition systems, this thesis focused on three major parts: first it discussed the importance of fingerprint enhancement before the recognition process. It showed that the enhancement process greatly improved the accuracy and security of the recognition process. Second, it presented a new approach to improve the performance of the fingerprint recognition system using image fusion techniques. Third, the thesis presented a decision fusion approach using the logical "OR" and logical "AND" operators.;Finally, this thesis presented an overall comparison of performance between traditional systems that used a single pattern and our proposed system that used two fused fingerprint patterns in the fingerprint recognition system by using the concept of receiver operator characteristic (ROC) curve. Analysis on the superiority of our approach is presented. |
