DeepIrisNet2: Learning Deep-IrisCodes from Scratch for Segmentation-Robust Visible Wavelength and Near Infrared Iris Recognition

TL;DR

DeepIrisNet2 introduces a deep learning framework for iris recognition that operates without traditional normalization or segmentation, using novel layers and strategies to handle non-ideal conditions and generate compact binary iris codes.

Contribution

The paper presents a new deep learning-based iris recognition framework that eliminates the need for classical normalization and segmentation, and introduces large labeled datasets for training.

Findings

Outperforms previous state-of-the-art methods on multiple datasets

Effectively handles non-ideal iris images without normalization

Generates compact binary iris codes for efficient recognition

Abstract

We first, introduce a deep learning based framework named as DeepIrisNet2 for visible spectrum and NIR Iris representation. The framework can work without classical iris normalization step or very accurate iris segmentation; allowing to work under non-ideal situation. The framework contains spatial transformer layers to handle deformation and supervision branches after certain intermediate layers to mitigate overfitting. In addition, we present a dual CNN iris segmentation pipeline comprising of a iris/pupil bounding boxes detection network and a semantic pixel-wise segmentation network. Furthermore, to get compact templates, we present a strategy to generate binary iris codes using DeepIrisNet2. Since, no ground truth dataset are available for CNN training for iris segmentation, We build large scale hand labeled datasets and make them public; i) iris, pupil bounding boxes, ii) labeled iris texture. The networks are evaluated on challenging ND-IRIS-0405, UBIRIS.v2, MICHE-I, and CASIA v4 Interval datasets. Proposed approach significantly improves the state-of-the-art and achieve outstanding performance surpassing all previous methods.