Design and Analysis of Deep-Learning Based Iris Recognition Technologies by Combination of U-Net and EfficientNet

In this paper, the effective deep-learning based methodology is developed for iris biometric authentication. Firstly, based on the U-Net model, the proposed system uses the semantic segmentation technology to localize and extract the region of interest (ROI) of iris. After the ROI of iris in the eye image is revealed, the inputted eye image will be cropped to the small-size eye image with the just-fitted ROI of iris. Then, the iris features of the cropped eye image are strengthened optionally by adaptive histogram equalization or Gabor filtering process. Finally, the cropped iris image is classified by the EfficientNet model. By the Chinese Academy of Sciences Institute of Automation (CASIA) v1 database, the proposed deep-learning based iris recognition scheme reaches the recognition accuracies up to 98%. Compared with the previous works, the proposed technology can provide the effective iris recognition accuracy for the biometrics applications with iris information.