Unconstrained Iris Segmentation Using Convolutional Neural Networks

Abstract

The extraction of consistent and identifiable features from an image of the human iris is known as iris recognition. Identifying which pixels belong to the iris, known as segmentation, is the first stage of iris recognition. Errors in segmentation propagate to later stages. Current segmentation approaches are tuned to specific environments. We propose using a convolution neural network for iris segmentation. Our algorithm is accurate when trained in a single environment and tested in multiple environments. Our network builds on the Mask R-CNN framework (He et al., ICCV 2017). Our approach segments faster than previous approaches including the Mask R-CNN network. Our network is accurate when trained on a single environment and tested with a different sensors (either visible light or near-infrared). Its accuracy degrades when trained with a visible light sensor and tested with a near-infrared sensor (and vice versa). A small amount of retraining of the visible light model (using a few samples from a near-infrared dataset) yields a tuned network accurate in both settings. For training and testing, this work uses the Casia v4 Interval, Notre Dame 0405, Ubiris v2, and IITD datasets.

A Robust Iris Segmentation Scheme Based on Improved U-Net

“…Compared with FCEDNs-bayesian-basic, the nice1, nice2, and f1 score of our method are improved by 3.33%, 2.82%, and by 5.44%, respectively. Also, our method is 1.56% better than Ahmad and Fuller [30] in f1 score, and our f1 score is similar to that of Shabab [17]. Compared with Yang et al, our method scores 0.37% lower in nice2, and the f1 score is also 1.24% lower, but the nice1 score is increased by 0.21%.…”

“…The proposed method is compared with the traditional algorithms [26]- [29] and the CNN iris segmentation methods [15]- [17], [30] on the premise of using the same dataset and the same tagged image data. Tables 2 and 3 show the segmentation results of these methods on the near-infrared illumination iris dataset; Table 4 shows the segmentation results of these methods on the visible light illumination iris dataset.…”

Iris Deidentification With High Visual Realism for Privacy Protection on Websites and Social Networks

“…To segment the iris from the ocular images I lef t , I right , we use an algorithm based on CNNs [38] because of its high accuracy in segmenting iris images acquired in visible light This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ This article has been accepted for publication in a future issue of this journal, but has not been fully edited.…”

Iris Recognition with Image Segmentation Employing Retrained Off-the-Shelf Deep Neural Networks

“…Ahmad and Fuller [1] used the same supervised learn-ing scenario to train the Mask R-CNN [13] for iris segmentation. The authors used CASIA-Iris-Interval-v4, Notre Dame 0405, and UBIRIS benchmarks for which the ground-truth segmentation masks were available.…”