An End to End Deep Neural Network for Iris Segmentation in Unconstraint Scenarios

With the increasing imaging and processing capabilities of today’s mobile devices, userauthentication using iris biometrics has become feasible. However, as the acquisition conditionsbecome more unconstrained and as image quality is typically lower than dedicated iris acquisitionsystems, the accurate segmentation of iris regions is crucial for these devices. In this work, an end toend Fully Convolutional Deep Neural Network (FCDNN) design is proposed to perform the irissegmentation task for lower-quality iris images. The network design process is explained in detail,and the resulting network is trained and tuned using several large public iris datasets. A set ofmethods to generate and augment suitable lower quality iris images from the high-quality publicdatabases are provided. The network is trained on Near InfraRed (NIR) images initially and latertuned on additional datasets derived from visible images. Comprehensive inter-databasecomparisons are provided together with results from a selection of experiments detailing the effectsof different tunings of the network. Finally, the proposed model is compared with SegNet-basic, anda near-optimal tuning of the network is compared to a selection of other state-of-art iris segmentationalgorithms. The results show very promising performance from the optimized Deep Neural Networksdesign when compared with state-of-art techniques applied to the same lower quality datasets.