Feature Extraction Classification for Personal Identification using Iris

This paper presents a new approach for feature extraction of iris image for recognition purpose. Two methods are used in this work for feature extraction and classification. Firstly input image is pre-processed where iris region and pupil region are extracted from the eye image using Canny edge detection and Hough Transform. Then key features are extracted from the iris region and using these features classification is performed for iris recognition. Combination of three feature extraction algorithms and a classifier is used in each method. The combination of Gabor filter, Principally Rotated complex wavelet filter (PR-CWF), Discrete Wavelet Transform (DWT) and Support Vector Machine (SVM) is used as method1. Similarly combination of PR-CWF, Bayer's Nearest Neighbour (BNN), Fast Fourier Transform and Artificial Neural Network (ANN) is used as method2.