Performance Analysis on High Efficient Score Level Fusion based Multimodal Biometric Person Recognition method for High Security Applications

Abstract

A biometric system that relies on the proof of a solitary source of information for personal recognition is able to deploy in real world application, which are said to be unimodal. The most regular single modal biometric framework can be seen in the greater part of the spots because of quality. It is unwavering quality has diminished on the grounds that it requires bigger memory foot shaped impression, higher operating expense and it has slower handling velocity. In this paper high effective Multimodal Biometric recognition scheme using Score level fusion is used which utilizes iris and unique finger impression for personal identification. The significant playing point of this multimodal methodology is that can put together the proof presented by multiple sources of information. In this paper discuss the different circumstances which are feasible in multimodal biometric systems, the levels of fusion that are plausible and the integration strategy can be adopted to strengthen information. Incorporating different modalities in client check, distinguishing proof prompts seem to be superior, high dependability and high precision. It upgrades safety for high security applications like border security, high secured internet managing, pilot identification in airplanes and war fields and hence saves life and belongings.

Keywords: Fingerprint Recognition, Iris Recognition, Multimodal Biometrics Identification, Score level fusion, Wavelet Transform, Neural Network.