Classification of Human Eye Image using Different Methods for Medical Applications

Abstract: The image classification is the method of assigning one or more category labels to an input image pattern. The Eye image classification aims to classify the iris and pupil to an application specific category as drug effect detection, left to right classification and fake classification. Iris texture is regarded as a genotypic biometric pattern which is stable during life time so that iris recognition and classification are reliable methods for individual authentication. The framework includes iris image pre-processing, low level feature extraction, statistical representation and iris image classification. The eye image database consists of eye images collected from different Persons and are used for testing the classification system. The Pupil is helpful in the detection of effect of drug. The Support Vector Machine Classifier is the well established classifier. It can be used to predict the class labels. In addition to this the classification is done by using total number of key points extracted and Diameter of the pupil.

Keywords: Support Vector Machine, Key points, Descriptors, Pin Point, Dilated.