Score Level Fusion for Iris and Periocular Biometrics Recogniton Based on Deep Learning

Abstract Traditional iris recognition has high accuracy and low misrecognition rate. However, in the case of mobile terminal or distance, image resolution quality decrease, rate also decreases. To solve above problems, this article is based on deep learning technology, basis single mode state recognition, from different levels multimodal integration, eyes score level fusion research, put forward adaptive dynamic weighted method, to determine weighing values can algorithm modal, without artificial specified, weighting more flexible, stronger applicability. Experimental results casIA-Iris-LAMP CasIA-Iris-Distance Iris database Chinese Academy Sciences show that proposed higher better performance than traditional fractional which proves effectiveness algorithm.