Implimentation of Fkp Based Biometric Identification System Using Pca with Neuro Fuzzy Neural Network

Biometric authentication is utilized in software engineering applications as a type of recognizable and access control of the particular person using their behavioral characteristics. Several biometric features such as fingerprint, palm veins, recognition, palm, hand geometry, iris recognition, DNA, and so on, used for authenticating the user identities. From the various biometric features, the finger-knuckle print (FKP) having the fine, rich texture, outer-palm surface is high also stable features which are difficult to hack by the intermediate person. In addition the FKP biometric feature is difficult to modify by the people emotional activity and other environment activities. So, the proposed system uses the FKP as the biometric feature while analyzing the authentication in the various applications. Initially the biometric FKP image is preprocessed by using the Gabor filter and the exact regions are segmented using the edge with region of interest method. From the extracted region different features are extracted with the help of a kernel and sparse principal component analysis. Finally the matching is performed with the help of the Neuro fuzzy neural network. The performance was tested with PolyU Finger Knuckle Printing database. The effectiveness of the proposed system in terms of False Accept Rate (FAR), False Rejection Rate (FRR), Equal Error Rate and Accuracy.