A Novel Classifier for Gender Classification from Iris Code used for Recognition

Abstract

-Biometric is the measurement and statistical analysis of human body characteristics. Biometric recognition is the use of biological measurements of an individual for identification purposes. Iris recognition system include image acquisition, Segmentation, normalization, features extraction, encoding and classification. Iris images are downloaded from CASIA Iris V1.0 database. Canny edge detection is used to detect the edges of the eye image. Hough Transform is used to separate the iris region from the eye image. Also to localize circular iris and pupil region of eye image, a circular Hough transform is used. Normalization technique convert the polar coordinated into Cartesian coordinates. These Normalized image is used for feature extraction. Feature extraction is performed by convolving the normalized iris region with Gabor wavelet, GLCM, etc. Depend on different features selection methods, find the feature set. Finally to classify the eye images using SVM and ANN classifiers depends on features extracted from the database images.