Feature Fusion Based Iris and Retina Recognition System Using Kohonon Self-Organizing Mapping Neural Network Algorithm

Abstract

This paper proposes a model of biometric security with feature fusion based iris and retina recognition system. Though, a lot of research works have done for iris recognition system and it is not new for the fusion of multimodal iris recognition in the area of biometric security and authentication system. But, it is a relatively new idea of mixing iris and retina features for human authentication. Here, the features of iris and retinal images are extracted using standard methods, then features are fused and reduced using Principal Component Analysis based dimensionality reduction technique. These reduced features are feed to the Kohonon Self-Organizing Neural Network Algorithm for learning and recognition purposes. CASIA iris dataset have been used for iris recognition and DRIVE dataset have been used for performance measurements of the retinal images.

Keywords- Biometric Security, Feature Fusion Model, Self-Organizing Mapping Neural Network Algorithm, Dimensionality Reduction Technique

Cite this Article

Md. Rabiul Islam, Md. Fayzur Rahman, Feature Fusion Based Iris and Retina Recognition System Using Kohonon Self-Organizing Mapping Neural Network Algorithm, Trends in Electrical Engineering (TEE). 2015; 5(2): 22–26p.