Experimental Evaluation of Secured IRIS Recognition System using Learning Assisted Classification and Prediction Methodology

An IRIS Recognition System is a biometric technology that identifies individuals based on the unique patterns in their irises, which are the colored part of the eye surrounding the pupil. This technology relies on the fact that every individual has distinct iris patterns, even between identical twins. In this study, we present an experimental evaluation of the Secured IRIS Recognition System (SIRISRS) employing a Learning Assisted Classification and Prediction Methodology. The core of our approach lies in the utilization of a Hybrid Inception and Convolutional Neural Network (CNN) model, a fusion of Inception architecture and traditional CNN.The SIRISRS demonstrates its proficiency in iris recognition, an area of biometrics vital for authentication and security applications. Leveraging a dataset with diverse iris patterns, our methodology focuses on feature extraction and classification. Through rigorous evaluation, we assess the system’s recognition accuracy and efficiency. Performance metrics such as Sensitivity, Accuracy, Specificity, Precision, Recall, and F Measure are employed to comprehensively gauge its capabilities. Our results underscore the SIRISRS’s exceptional accuracy, cementing its position as a formidable solution for iris recognition in security and authentication scenarios, which showcases remarkable accuracy, achieving a rate of 99.85%. This study contributes valuable insights into the fusion of advanced neural network architectures and machine learning in biometric systems, opening avenues for enhanced security and identity verification applications.