DETECTION OF DIABETIC PRESENCE FROM IRIS BY USING S UPPORT VECTOR MACHINE

This is a simple and non - intrus ive method to detect diabetic in body and iris recognition is not only mainly for biometric identification but it can also be used as a mean to detect diabetic or maybe diagnose any diseases as iridology claimed it is supposed to be. The attempt is being made to explore the area of dia gnosis from different perspectives. The approach used is a combination of ancestor's technology Iridodiagnosis with modern technology. To begin with a database is created of eye images with clinical history of subject’s emphasis on diabetic (type II) dise a se in pathological laboratory. The various algorithms are developed for image quality assessment, segmentation of iris, iris normalization and clinical feature classification for clinical diagnosis. The Support Vector Machine is used for trainin g and classification purpose. Others application that can use this program is to determine the eye problem due to other type of eye diseases such as cataract, glaucoma, cholesterol , tumour etcetera. This approach will be useful in the diagnosis field whic h is faster, user friendly and less time consuming