International Conference on Innovations in Information, Embedded and Communication Systems |
Foundation of Computer Science USA |
ICIIECS - Number 1 |
November 2014 |
Authors: S S Tamilchelvi, N B Mahesh Kumar |
2e0ef175-1dcd-4d15-a589-ae6b7be207d2 |
S S Tamilchelvi, N B Mahesh Kumar . Pamlprint Biometric System for Personal Authentication. International Conference on Innovations in Information, Embedded and Communication Systems. ICIIECS, 1 (November 2014), 14-17.
Biometrics refers to technologies that measure and analyzes human body behavioral characteristics such as fingerprints, palmprint, knuckles, hand geometry hand vein, retinas, irises and physiological characteristics such as gait, signature, key stroke etc. for authentication purposes. The palmprint is emerging biometric with the advantages of large Region of Interest (ROI) and chances of damages are very rare. In biometric based authentication decision tree is used for the classification of genuine and imposter classes. If the data is noisy, decision tree leads to false branches called over-fitting of the tree. This requires pruning of tree which will lead to decrease in performance. For that Fuzzy Decision Trees (FDT) is developed for classification of claimed identity into two classes genuine and imposter, hence it termed as Fuzzy Binary Decision Tree (FBDT). Fuzzy gini index and fuzzy entropy is used for selection of tree nodes. Matching training scores are used for automatic generation of fuzzy membership functions.