Research on Human Eye Iris Recognition Based on Inception v3
Yuewen Tang, Xiangkui Li, Yulian Jiang · Frontiers in Signal Processing · 2019
Iris recognition has certain application prospects in access control systems, financial fields, and airport security.Before the appearance of the deep learning method, the iris recognition method relies on artificial method for feature extraction.This method will consume a lot of time and affect the efficiency of recognition.In this recognition study, we adopt the method of deep learning.On the basis of a large amount of image data and the guarantee of computer computing ability, in order to exceed the iris recognition accuracy of previous recognition methods.In this paper, the improved Inception v3 model is used to classify and recognize the iris database provided by the Chinese Academy of Sciences in the Tensorflow framework.Finally, the accuracy of image recognition can reach 98% through the training of the model.