Advanced Iris Recognition: A Deep Learning Approach to Human Identification Using Convolutional Neural Networks
Rekha Sachin Kamble1, Varsha Ravsaheb Kamble2, Shrenik R. Patil3
1Assistant Professor, DKTE Society's Textile & Engineering Institute (An Empowered Autonomous Institute), Ichalkaranji
2M. Tech Scholar, DKTE Society's Textile & Engineering Institute (An Empowered Autonomous Institute), Ichalkaranji
31Assistant Professor, DKTE Society's Textile & Engineering Institute (An Empowered Autonomous Institute), Ichalkaranji
--------------------------------------------------------------------***-------------------------------------------------------------------
Abstract - The component of a computer system that is in charge of user security is one of the most crucial ones. Simple logins and passwords have been shown to be vulnerable to hackers and unable to provide high levels of efficiency. The most common replacements is identity recognition using biometrics. The use of the iris as a biometric feature has gained popularity in recent years. It resulted from the exceptional efficiency and precision that this approach provided. The results of this interest may be seen in the literature. Various authors have put forward a variety of various methods. The authors of this paper describe their own method for an iris-based algorithm for recognizing human identification. Artificial neural networks and a CNN-based transfer learning model (Mobile Net) were employed in the classification process. As soon as the classification is complete, the iris section is segmented on the result of the classification. The proposed approach can produce results that are adequate, according to tests that have been run.
Key Words: Iris-based human identity recognition, CNN, Transfer learning, Image segmentation, artificial neural networks
