Finger Knuckle Print based Biometric System Creation using Dual Clustering with Radial Basis-Manhattan Length Approach

S. Suganthi Devi
2021 International Journal for Research in Applied Science and Engineering Technology  
Authentication is important factor in most of the application to perform the verification process. There are several factors are used to verify the personal identifies but biometric characteristics are most important one because everyone are having specific biometric features. Due to the easy access, minimum cost, stable features are leads to chose the finger knuckle print for authentication purpose. Initially, finger knuckle print images are collected from people which are processed by
more » ... rocessed by applying mean filter. Then the finger knuckle region is located by applying the optimized dual clustering approach. Then the various features are extracted which are trained with the help of the sigmoid activation function with radial neural network. Finally, the extracted features are matched with the trained feature using Manhattan length. This described matching process helps to authenticate the users. At last efficiency of the system is evaluated using MATLAB based experimental results such as false acceptance rate, equal error rate and false rejection rate.
doi:10.22214/ijraset.2021.34596 fatcat:v36vl3dlmndobndamkz4cpvtde