KCIR: A Novel Iris Recognition System using Deep CNN with Kalman Filtering

In today’s environment, authorizing somebody has become a critical requirement. In such conditions, the incorporation of artificial intelligence into biometric authentication systems has altered the lives of people and operations at diverse stages. Machine learning is becoming ever more popular in various fields of computer science. A powerful visual representation of machine learning is a deep conventional neural network. For an authentication system using iris, it tends to present resilience and effective structure. The Iris arrangement is a biological trait that is unique to each person, paving way for a vital and effective tool for authenticating a person. This research provides a robust iris recognition strategy based on a Convolutional Neural Network using Kalman Filter. The suggested system outperforms certain current iris recognition strategies on public iris databases, such as Ubiris.v2, CASIA, and MMU V1.0, in terms of experimental findings, with an accuracy of above 99 percent.