skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Blink detection for off-angle iris images using deep learning
Iris recognition is one of the well-known areas of biometric research. However, in real-world scenarios, subjects may not always provide fully open eyes, which can negatively impact the performance of existing systems. Therefore, the detection of blinking eyes in iris images is crucial to ensure reliable biometric data. In this paper, we propose a deep learning-based method using a convolutional neural network to classify blinking eyes in off-angle iris images into four different categories: fully-blinked, half-blinked, half-opened, and fully-opened. The dataset used in our experiments includes 6500 images of 113 subjects and contains images of a mixture of both frontal and off-angle views of the eyes from -50 to 50 in gaze angle. We train and test our approach using both frontal and off-angle images and achieve high classification performance for both types of images. Compared to training the network with only frontal images, our approach shows significantly better performance when tested on off-angle images. These findings suggest that training the model with a more diverse set of off-angle images can improve its performance for off-angle blink detection, which is crucial for real-world applications where the iris images are often captured at different angles. Overall, the deep learning-based blink detection method can be used as a standalone algorithm or integrated into existing standoff biometrics frameworks to improve their accuracy and reliability, particularly in scenarios where subjects may blink.  more » « less
Award ID(s):
2100483
PAR ID:
10431642
Author(s) / Creator(s):
; ;
Editor(s):
Alam, Mohammad S.; Asari, Vijayan K.
Date Published:
Journal Name:
Pattern Recognition and Tracking XXXIV
Volume:
1252707
Page Range / eLocation ID:
6
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, IEEE)
    Iris recognition is known to be a robust biometric method, leveraging distinctive iris patterns for reliable identification. However, challenges arise in traditional approaches, especially in non-ideal images captured in environments where subjects can freely move around. This research focuses on the impact of eyelid occlusion on off-angle iris recognition within a traditional framework. The primary objectives are to i) assess the impact of eye blink on overall recognition performance; ii) explore the effects on each specific angle; and iii) examine the occlusion level required to maintain an acceptable recognition performance in a traditional iris recognition pipeline. To generate different levels of eyelid occlusion, we manipulated the eyelid segmentation results to mimic eye blinking due to lack of adequate amount of eye blinking images in publicly available datasets. We conducted three sets of experiments using frontal and off-angle iris images from 100 different subjects. Based on the results, up to a certain occlusion level (60%), eyelid occlusion enhances performance for severe off-angle images by masking distortion-prone iris portions. This improvement comes from the masking of non-ideal portions of the off-angle iris images that are distorted by the refraction of light through the cornea caused by the gaze angle. This insight offers potential improvements for traditional iris recognition, highlighting the nuanced interplay of occlusion and gaze angle on recognition performance. 
    more » « less
  2. ; (, Multimodal Image Exploitation and Learning)
    Agaian, Sos S.; DelMarco, Stephen P.; Asari, Vijayan K. (Ed.)
    Iris recognition is a widely used biometric technology that has high accuracy and reliability in well-controlled environments. However, the recognition accuracy can significantly degrade in non-ideal scenarios, such as off-angle iris images. To address these challenges, deep learning frameworks have been proposed to identify subjects through their off-angle iris images. Traditional CNN-based iris recognition systems train a single deep network using multiple off-angle iris image of the same subject to extract the gaze invariant features and test incoming off-angle images with this single network to classify it into same subject class. In another approach, multiple shallow networks are trained for each gaze angle that will be the experts for specific gaze angles. When testing an off-angle iris image, we first estimate the gaze angle and feed the probe image to its corresponding network for recognition. In this paper, we present an analysis of the performance of both single and multimodal deep learning frameworks to identify subjects through their off-angle iris images. Specifically, we compare the performance of a single AlexNet with multiple SqueezeNet models. SqueezeNet is a variation of the AlexNet that uses 50x fewer parameters and is optimized for devices with limited computational resources. Multi-model approach using multiple shallow networks, where each network is an expert for a specific gaze angle. Our experiments are conducted on an off-angle iris dataset consisting of 100 subjects captured at 10-degree intervals between -50 to +50 degrees. The results indicate that angles that are more distant from the trained angles have lower model accuracy than the angles that are closer to the trained gaze angle. Our findings suggest that the use of SqueezeNet, which requires fewer parameters than AlexNet, can enable iris recognition on devices with limited computational resources while maintaining accuracy. Overall, the results of this study can contribute to the development of more robust iris recognition systems that can perform well in non-ideal scenarios. 
    more » « less
  3. ; (, IEEE)
    Iris biometric systems offer non-contact authentication, particularly advantageous in controlled environments such as security checkpoints. However, challenges arise in less controlled scenarios such as standoff biometrics where captured images mostly are non-ideal including off-angle. This paper addresses the need for iris recognition models adaptable to various gaze angles by proposing a blink detection algorithm as an additional feature. The study explores different blink detection methods including involving logistic regression, random forest, and deep learning models. For the first methodology, logistic regression and a random forest model were used to classify eye images into four different blink classes. The second methodology involved labeling eye openness percentage. The ground-truth eye blink was calculated using facial landmarks detected by the MediaPipe model. For the deep learning approach, we used a pre-trained Convolutional Neural Network (CNN) model by replacing the output layer with a regression layer. Results show improved precision and recall when incorporating height and width features for the regression model. The AlexNet model achieves superior performance, reaching 90% accuracy with a 10 % error threshold. This research contributes valuable insights for developing robust iris recognition models adaptable to diverse gaze angles. 
    more » « less
  4. ; ; (, ACM)
    This study explores the impact of blinking on deep learning based iris recognition, addressing a critical aspect in the development of robust, reliable, and non-intrusive biometric systems. While previous research has demonstrated the promise of Convolutional Neural Networks (CNNs), such as AlexNet, GoogleLeNet, and ResNet, the impact of blinking remains underexplored in this context. To address this gap, our research focuses on training multiple ResNet models with varying degrees of iris occlusion exposure. Using a dataset with 101 subjects, we generated cohorts of synthetically occluded images ranging from 0% occlusion to 90% occlusion. Our findings reveal a noteworthy linear performance decrease in models unexposed to blinked images as iris occlusion increases. However, augmenting the training dataset with occluded images significantly mitigates this performance degradation, highlighting the importance of accounting for blinking in the development of reliable iris recognition systems. 
    more » « less
  5. (, 2021 IEEE International Workshop on Biometrics and Forensics)
    null (Ed.)
    As an emerging biometric research, standoff iris recognition systems focus on recognition of non-cooperative subjects in much less constrained environments where their captured images are likely to be non-ideal including being off-angle. Iris biometrics convert unwrapped iris textures into binary iris codes to compare them with other saved codes by measuring their Hamming Distances. The similarity calculation assumes an equal contribution of each individual pixel in iris codes. However, previous studies showed that some pixels (aka. fragile bits) are more error prone than others even in frontal iris images. In addition, off-angle iris images are affected by several challenging factors including corneal refraction and limbus occlusion. These challenges in off-angle images also increase the fragility of bits in iris codes. This paper first presents the pixel inconsistency in iris codes of off-angle images using elliptical segmentation and normalization. The pixel fragility is a result of iris codes warping due to the refraction of light in cornea and occlusion of iris texture at limbus. As another contribution, we propose to identify these fragile pixels in iris codes using edge detection and eliminating them in Hamming distance calculation by masking these fragile bits. Based on the results, the proposed method improves the recognition performance in off-angle iris images where the average genuine Hamming distance score reduced from 0.3082 to 0.1244 and the equal error rate is lowered 19%. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email