A deep learning based unified framework to detect, segment and recognize irises using spatially corresponding features. (September 2019)

• Record Type: Journal Article
• Authors: Zhao, Zijing; Kumar, Ajay
• Is Part Of: Pattern recognition. Volume 93(2019:Sep.)
• Page Start: 546
• Page End: 557
• Publication Date: 2019-09
• Subjects: Iris recognition -- Deep learning -- Spatially corresponding features; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.04.010
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: A novel and unified deep learning based framework to automatically detect, segment and recognize irises from eye images. Higher iris detection and segmentation accuracy over previous approach. Outperforming results over other methods for iris recognition in the literature (e.g., ICCV 2017) using publicly accessible databases. Comparative experimental results using a range of promising iris recognition methods. Abstract: This paper proposes a deep learning based unified and generalizable framework for accurate iris detection, segmentation and recognition. The proposed framework firstly exploits state-of-the-art and iris-specific Mask R-CNN, which performs highly reliable iris detection and primary segmentation i.e., identifying iris/non-iris pixels, followed by adopting an optimized fully convolutional network (FCN), which generates spatially corresponding iris feature descriptors. A specially designed Extended Triplet Loss (ETL) function is presented to incorporate the bit-shifting and non-iris masking, which are found necessary for learning meaningful and discriminative spatial iris features. Thorough experiments on four publicly available databases suggest that the proposed framework consistently outperforms several classic and state-of-the-art iris recognition approaches. More importantly, our model exhibits superior generalization capability as, unlike popular methods in the literature, it does not essentially require database-specific parameter tuning, which is another key advantage.