Laser & Optoelectronics Progress, Volume. 62, Issue 8, 0800001(2025)
Application Prospects of Iris Recognition Technology in Identity Authentication for Standalone Virtual Reality Devices
Figures & Tables(20)
![Iris images under different illuminations[28]. (a) Iris images collected under visible light; (b) iris images collected under infrared light](/Images/icon/loading.gif){kind=icon}
Fig. 4. Iris images under different illuminations[28]. (a) Iris images collected under visible light; (b) iris images collected under infrared light
Fig. 5. Iris images under different illuminations[31]. (a) Iris images under four 765 nm LEDs; (b) iris images under two 765 nm LEDs and one 850 nm LED; (c) iris images under two 850 nm LEDs; (d)‒(f) corresponding segmented iris images
Fig. 8. Iris recognition equipments[35]. (a) Close iris recognition equipment; (b) (c) medium and long-distance recognition equipments
Fig. 10. Poor quality iris images[44]. (a) Eyelid occlusion; (b) closed eyes; (c) eyelash occlusion; (d) iris is not clear
Fig. 11. Iris images with different resolutions[48]. (a) High resolution iris image; (b) low resolution iris image
Fig. 12. Iris localization algorithm[78]. (a) Localization result of Daugman algorithm; (b) localization result of Wildes algorithm
Fig. 16. Camera schematic diagram[104]. (a) Traditional three-chip lens camera; (b) metalenses camera
Table 1. Feasibility analysis of four biometric technologies for implementing identity authentication function in standalone VR headsets
View tableTable 1. Feasibility analysis of four biometric technologies for implementing identity authentication function in standalone VR headsets
Bio identification technology Security class Learning cost Structural change Existing interference Fingerprint recognition High High Fingerprint recognition module； collection plane for fingerprint images Physical contact； higher fingerprint clarity and integrity Face recognition Higher Low Face recognition module； more cameras for facial features Unable to collect complete face information Iris recognition Highest Middle Eye trace module for iris recognition Closing eyes； strabismus； obstruction； mirror reflection； ophthalmology diseases Speech recognition Higher Middle No need to change Physiology； pathology； camouflage； environment
Table 2. Image deblurring
View tableTable 2. Image deblurring
Classification Ref. Approach Pros and Cons Traditional method Non-blind deblurring ［ 50 ］Lucy-Richardson algorithm Simple model structure， but limited adaptation scene ［ 51 ］Wiener filtering algorithm Blind deblurring ［ 52 ］Maximum a posterior Improved application scenario， but limited resilience in complex scenarios ［ 53 ］Variational Bayesian Deep learning method ［ 54 ］Kernel estimation Stronger fuzzy kernel estimation ability ［ 55 ］End-to-end network Improved accuracy of the recovery effect， but long training time ［ 56 ］Dynamic network Improved deblurring performance
Table 3. Super-resolution reconstruction of images
View tableTable 3. Super-resolution reconstruction of images
Classification Ref. Approach Pros and cons Interpolation algorithm ［ 57 ］Nearest neighbor interpolation Simple and fast， but poor effect with blurred edge ［ 58 ］Bilinear interpolation ［ 59 ］Bicubic interpolation Reconstruction algorithm ［ 60 ］Iterative back projection Sharper edge， but degrade rapidly for HR image with large-scale factor problem ［ 61 ］Projection onto convex set ［ 62 ］Maximum a posteriori Shallow learning algorithm ［ 63 ］Example-based methods Fine high frequency information depending on the quality of the training set ［ 64 ］Manifold More image details can be restored ［ 65 ］Compressive sensing High quality of the reconstructed image， but complex algorithm Deep learning algorithm ［ 66 ］［67 ］CNN Automatic extraction and high semantic level of image feature information with deeper network level and difficult training ［ 68 ］Residual network （ResNet） Improve CNN’s gradient disappearance and gradient explosion problem ［ 69 ］Generative adversarial networks （GAN） Reconstruct super resolution images with texture details for large scaling factor
Table 4. Comparison of traditional iris feature extraction methods
View tableTable 4. Comparison of traditional iris feature extraction methods
Approach Recognition effect Robustness of illumination Ability to resist noise interference LBP High Higher Low Based on Haar wavelet transform High Higher Low Based on two-dimensional Gabor filtering Medium Low Medium GLCM Low High High
Tools
Get Citation
Copy Citation Text
Zhenyu Ma, Nini Wang, Guolei Wu, Chenlong Zhu, Mingqi Zhou, Yongjun Liu, Yunxiang Yan. Application Prospects of Iris Recognition Technology in Identity Authentication for Standalone Virtual Reality Devices[J]. Laser & Optoelectronics Progress, 2025, 62(8): 0800001
Paper Information
Category: Reviews
Received: May. 22, 2024
Accepted: Sep. 24, 2024
Published Online: Apr. 3, 2025
The Author Email: Yunxiang Yan (yanyunxiang@hrbeu.edu.cn)
CSTR:32186.14.LOP241341
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20