Implementation of Fast and Reliable Multi Spectral Iris Segmentation Using Deep Learning

Abstract

Iris recognition techniques have gained significant attention in the field of identification due to their high performance and robustness. The unique texture of the iris makes it an ideal biometric for individual identification. However, there are various challenges that arise in real-world recognition scenarios. This thesis focuses on different phases of the iris recognition system and reviews the methods associated with each phase. newlineThe recognition system is composed of seven phases: acquisition, preprocessing, segmentation, normalization, feature extraction, feature selection, and classification. The thesis also discusses two approaches to iris recognition: the traditional approach and the deep learning approach. newlineIris segmentation is a critical step in iris recognition systems and has a direct impact on authentication and recognition results. However, standard segmentation techniques may not perform well in noisy iris databases captured under challenging conditions. Moreover, the lack of large iris databases hinders the performance improvement of convolutional neural networks. The proposed method addresses these challenges by effectively handling irregular iris images captured under visible light. The iris region is processed and evaluated to generate a unique feature vector, which is then used for person identification. VGG16, a well-known deep learning model, is employed for image classification. The model is trained on the CASIA-Iris-Dataset, and the experimental results demonstrate its superiority over existing methods, achieving an accuracy of up to 90.91% for the VGG16 Deep Iris models.