Development of Iris Biological Features Extraction for Biometric Based Authentication to Prevent Covid Spread

Abstract

A biometric based authentication is mainly used in the automated security systems. The contact biometrics leads to spread the infections like corona. The proposed method uses non contact biometric sources. In this method an analysis is made on the iris pigment spot, crypt, and wolflin nodules biological features. The features are individually segmented in the rectangular form. The BRISK, SURF, FAST, MinEigen, MSER, and Harris features are extracted separately from the Wolflin nodule, crypt, and Pigment spot of the iris region.  An analysis is carried with the number of feature points extracted from each methodology. The suitable feature extraction method for pigment spot, crypt, and wolflin nodules are identified based on the statistical result analysis. The 60% of the input is divided into trained and remaining 40% into untrained category.  A feature matching is performed with test and trained feature. The analysis is performed on number of feature matching. The authentication threshold level of pigment spot, crypt, and wolflin nodule are above 3, 2, and 0 respectively.