Abstract
Today, iris recognition becomes one of the most popular and useful techniques used for person verification, which takes advantage of the fact that the eye has some unique characteristics represented by the iris or more precisely in the pupil. Iris scanning biometric systems measure the unique patterns in the iris of human's eye to verify and authenticate their identity. Fast, contactless and renowned for its accuracy, biometric iris recognition can work at long distance with some solutions that leverage the modality requiring only a glance from a user. This paper presents a study on the performance of iris recognition system using PYTHON for implementation and CASIA-V I iris databases as training and testing images where seven hundreds of iris images were used. The measurements demonstrated our model provides higher identification accuracy than the VGG16 model