Abstract
The human finger nails have a high degree of distinctiveness, even in the case of identical twins or even between different finger nails of an individual. This is used as a key to develop a new biometric authentication system using a single nail plate. The nail authentication system developed is based on the high individuality of the dermal structure underneath the fingernail plate, known as nail bed. Because of inconsistent properties of growing nail plate the nail bed alone is considered. Semantic points mediation technique is used to form a pentagon structure on the nail bed and texture properties inside the structure is masked and used as Region of Interest (ROI). Principal Component Analysis (PCA), Independent Component Analysis (ICA), Haar Wavelet and Scale Invariant Feature Transform (SIFT) are used for feature extraction. Classification is done using naive Bayes and Support Vector Machine (SVM). The performance for each of the feature extraction algorithm with two classifiers is analyzed.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have