Decision Fusion based Pair of Iris Recognition using Back-Propagation Learning Neural Network Algorithm

Abstract

The The contribution of this work is to enhance the performance of the iris recognition system through decision fusion of left and right iris pattern. Iris recognition system performs well and identify human correctly in neutral environment. In this paper a pair of iris recognition system has been proposed, which is capable enough to identify human through noisy environments. Principal component analysis based dimensionality reduction technique has been used to reduce and extract the iris features from the left and right iris. Back-propagation learning neural network based learning and classification model has been applied into each left and right iris pattern separately and feature fusion based left-right iris feature. Three different recognition outputs are combined through various types of voting approach such as majority vote, maximum vote, average vote and nash vote to find out the final result. Experimental results show the superiority of the proposed decision fusion based pair of iris recognition system compared with existing individual left and right iris recognition system and feature fusion based left and right iris feature.

Keywords: Iris recognition, iris feature extraction, left-right feature fusion, principal component analysis, back-propagation neural network, decision fusion

Cite this Article Md. Rabiul Islam. Decision Fusion based Pair of Iris Recognition using Back-Propagation Learning Neural Network Algorithm. Journal of Computer Technology Application. 2015; 6(2): 1–6p.