Cross Sensor Multibiometric Authentication System

Precise identification of claimed identity is very important to the operation of our increasingly electronically interconnected information society. As a rapidly evolving technology Biometrics has been widely used in forensics, such as identifying criminals and prison security, and has the good potential to be adopted in a very broad range of civilian applications. The proposed system is trained using neural network (NN) algorithm and support vector machine (SVM) to achieve sensor adaptability. The result of training gives the adaptive parameters which will be stored into database. These stored adaptive parameters will be used at the time of matching for verification. The output of the system will be whether to accept the claimed identity or to reject