ISUR: Iris Segmentation based on UNet and ResNet

Iris segmentation is an important research subject. There are many studies for iris segmentation. Deep learning-based methods have improved iris segmentation accuracy greatly. Now, the main challenge for researchers is to segment the noisy iris images captured in non-ideal environments created by visible light and user non-cooperation. This study proposes a robust method based on deep learning to segment noisy iris images called ISUR that includes two stages: detection and segmentation. First, an attention mask is created via a Convolutional Neural Network (CNN) based detector. Second, the eye image is segmented using a new CNN, and the obtained attention mask is used to focus on the iris region in this network. The proposed method has been tested on well-known iris datasets: UBIRIS and Iris Distance subset from CASIA, and has shown promising results compared to UNet.