An Advance Facial Biometric System-Based Iris Classification with Image Processing and Deep Learning

Recent interest has been generated in multimodal biometrics technology due to its potential to increase recognition rates by overcoming some of the fundamental limitations of single biometric modalities. A typical biometric recognition system will consist of components for sensing, feature extraction, and matching. The system’s robustness is dependent on the accuracy with which relevant data can be gathered from certain biometric features. This study presents a novel face-iris trait feature extraction approach for use in multimodal biometric systems. Among the state-of-the-art algorithms for biometric based iris/face recognition, image processing algorithms with deep learning will have an edge as they are very solid. The iris authentication fits the complex mathematical patterns of the irises which are drastically particular for each. A comprehensive look on biometric authentication located that the fake rejection price of iris authentication is only 1.8% which is the bottom and highest accuracy of 93.78%.