Comparative Analysis of IRIS based Human Identity recognition using various Classification Algorithms

Abstract

The module responsible for user safety is one of the most vital components of computer systems. It has been shown that simple passwords and logins cannot ensure great efficiency and are simple for hackers to get. The well-known alternative is biometric identity recognition. In recent years, iris as a biometrics attribute has garnered more attention. This was owing to the great efficiency and precision assured by this quantifiable characteristic. In the literature, the effects of this curiosity may be found. Several diverse ways have been offered by various writers. Neither employs discrete fast Fourier transform (DFFT) components to characterise the iris sample. In this paper, the authors offer their unique method for iris-based human identification recognition using DFFT components determined via principal component analysis. Three techniques were utilised for classification: k-nearest neighbours, support vector machines, and artificial neural networks. Tests conducted have shown that the suggested procedure may provide good results.