The Impact of Preprocessing on Deep Representations for Iris Recognition on Unconstrained Environments
This addresses the problem of noisy iris recognition for biometric systems, though it appears incremental as it builds on existing deep learning architectures with domain-specific adaptations.
The paper tackles iris recognition in unconstrained environments by proposing deep representations using VGG and ResNet-50 networks without normalization, achieving a new state-of-the-art average EER of 13.98% on the NICE.II competition dataset, representing an absolute reduction of about 5%.
The use of iris as a biometric trait is widely used because of its high level of distinction and uniqueness. Nowadays, one of the major research challenges relies on the recognition of iris images obtained in visible spectrum under unconstrained environments. In this scenario, the acquired iris are affected by capture distance, rotation, blur, motion blur, low contrast and specular reflection, creating noises that disturb the iris recognition systems. Besides delineating the iris region, usually preprocessing techniques such as normalization and segmentation of noisy iris images are employed to minimize these problems. But these techniques inevitably run into some errors. In this context, we propose the use of deep representations, more specifically, architectures based on VGG and ResNet-50 networks, for dealing with the images using (and not) iris segmentation and normalization. We use transfer learning from the face domain and also propose a specific data augmentation technique for iris images. Our results show that the approach using non-normalized and only circle-delimited iris images reaches a new state of the art in the official protocol of the NICE.II competition, a subset of the UBIRIS database, one of the most challenging databases on unconstrained environments, reporting an average Equal Error Rate (EER) of 13.98% which represents an absolute reduction of about 5%.