Blink detection for off-angle iris images using deep learning

Blink detection for off-angle iris images using deep learning
复制标题

使用深度学习对斜角虹膜图像进行眨眼检测

DOI:
10.1117/12.2662248
复制
发表时间:
2023
期刊:
Pattern Recognition and Tracking XXXIV
影响因子:
--
通讯作者:
Karakaya, Mahmut
Karakaya, Mahmut
中科院分区:
--
文献类型:
--
作者:
Palta, Hasan;Omoteso, Timi;Karakaya, Mahmut

文献摘要

参考文献

被引文献

相似文献

虹膜识别是生物识别研究的一个著名领域。然而,在现实世界的场景中,受试者可能并不总是提供完全睁开的眼睛,这可能对现有系统的性能产生负面影响。因此,虹膜图像中眨眼的检测对于确保可靠的生物特征数据至关重要。在本文中,我们提出了一种基于深度学习的方法,使用卷积神经网络将倾斜虹膜图像中的眨眼分为四种不同的类别:完全眨眼,半眨眼,半睁开和完全睁开。在我们的实验中使用的数据集包括6500个图像的113个主题,并包含图像的正面和偏离角度的眼睛从-50o到50 o的注视角度的看法的混合物。我们使用正面和偏角图像训练和测试我们的方法,并对这两种类型的图像实现了高分类性能。与仅使用正面图像训练网络相比,我们的方法在对偏角图像进行测试时表现出更好的性能。这些发现表明,使用更多样化的偏角图像集训练模型可以提高其偏角眨眼检测的性能,这对于虹膜图像通常以不同角度捕获的现实应用至关重要。总的来说,基于深度学习的眨眼检测方法可以用作独立算法,也可以集成到现有的对峙生物识别框架中,以提高其准确性和可靠性,特别是在受试者可能眨眼的情况下。
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 -50o to 50o 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.
DOI: --
发表时间: 2009
期刊: --
影响因子: --
作者:
M. Divjak;H. Bischof
通讯作者: M. Divjak;H. Bischof
DOI: 10.11591/ijeecs.v22.i1.pp222-231
发表时间: 2021-04
影响因子: --
作者:
Pothuraju Vishesh;S. Raghavendra;SantoshKumar Jankatti;Rekha
通讯作者: Pothuraju Vishesh;S. Raghavendra;SantoshKumar Jankatti;Rekha
实时眨眼和眨眼检测,用于 HCI 系统中的对象选择
DOI: --
发表时间: 2018
影响因子: 2.9
作者:
Hari Singh Dhillon;J. Singh
通讯作者: J. Singh
DOI: 10.1109/icsipa.2011.6144085
发表时间: 2011-11
期刊: 2011 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)
影响因子: --
作者:
Axel Panning;A. Al-Hamadi;B. Michaelis
通讯作者: Axel Panning;A. Al-Hamadi;B. Michaelis
一种实时眨眼检测方法及其应用
DOI: --
发表时间: 2009
期刊:
影响因子: --
作者:
T. Srinark
通讯作者: T. Srinark