Iris Recognition using Convolutional Neural Network
Md. Shafiul Azam, Humayan Kabir Rana
Abstract
It is indispensable to ensure security biometrically in the most authentication and identification scenario. Iris recognition regarded as the most reliable biometric recognition due to its stable and extraordinary variation in texture. The unique patterns are used in iris recognition to identify individuals in requiring a high level of security. This paper explores an efficient technique that uses convolutional neural network (CNN) and support vector machine (SVM) for feature extraction and classification respectively to increase the efficiency of recognition. The proposed technique has been successfully applied and also clearly demonstrates the performance of the experimental evaluation on iris images from the CASIA database.