Comparative Analysis of Feature Selection Techniques For Fingerprint Recognition Based on Artificial Bee Colony and Teaching Learning Based Optimization

Abstract

Fingerprint biometric contains a tremendous number of textural elements which make the accurate fingerprint pattern classification challenging. Hence, an efficient algorithm for enhance recognition of fingerprint pattern is highly required. Various researches have revealed that feature selection techniques can be used to improve the discriminating ability and the computational burden of classifiers in the classification of fingerprint features. In this study, a comparative analysis of feature selection techniques for fingerprint recognition based on Artificial Bee Colony (ABC) and Teaching Learning Based Optimization (TLBO). The need to compare the performance of the two optimization techniques becomes necessary due to the fact that theoretical foundation of metaheuristic search algorithms suggested that no single algorithm is suitable for all problems. Hence, the better feature selection technique between the TLBO and ABC technique in fingerprint recognition system was investigated. Experimental result revealed that the TLBO technique outperformed the ABC technique and would produce a more reliable and accurate fingerprint authentication system.