ABSTRACT
Iris recognition for at-a-distance acquired iris images under less constrained environment has shown to be challenging due to highly imaging variations such as reflections, motion blur, occlusions etc. This poses challenges for conventional gradient-based iris segmentation methods which are essentially developed to work on high quality iris images acquired in a controlled environment. In this work, we propose an effective encoder-decoder Deep Convolutional Neural Network which can be trained end-to-end to perform iris segmentation for distantly acquired iris/face images. More specifically, the proposed approach is motivated by the recent state-of-the-art semantic segmentation approach -- DeepLabv3/3+. The encoder module adapts the ResNet-50 as base network and extended with additional blocks constructed using multi-grid atrous convolution, and Atrous Spatial Pyramid Pooling to capture multi-scale features, which can better accommodate the segmentation of iris at different scales. To facilitate recovering of the spatial information, refinement module is introduced in the decoder module. We demonstrate the effectiveness of the proposed approach on two public datasets, i.e., UBIRIS.v2 and FRGC, which achieves average improvement of 37.42% and 48.9%, respectively. The trained model is made publicly available at https://gitlab.com/cwtan501/iris_segmentation to encourage reproducible of the reported results.
- Aadhaar Authentication: 2016. https://www.uidai.gov.in/authentication/authentication-overview/authentication-en.html.Google Scholar
- Bowyer, K.W. et al. 2008. Image understanding for iris biometrics: A survey. Computer Vision and Image Understanding. 110, 2 (2008), 281--307. DOI:https://doi.org/10.1016/j.cviu.2007.08.005.Google ScholarDigital Library
- Daugman, J. 2015. Iris Recognition at Airports and Border Crossings. Encyclopedia of Biometrics. 998--1004.Google Scholar
- Jain, A.K. and Kumar, A. 2012. Biometric Recognition: An Overview. 49--79.Google Scholar
- Wildes, R.P. 1997. Iris recognition: An emerging biometrie technology. Proceedings of the IEEE. 85, 9 (1997), 1348--1363. DOI:https://doi.org/10.1109/5.628669.Google ScholarCross Ref
- Soft Computing and Image Analysis Lab: http://socia-lab.di.ubi.pt/.Google Scholar
- Proença, H. et al. 2010. The UBIRIS.v2: A database of visible wavelength iris images captured on-the-move and at-a-distance. IEEE Transactions on Pattern Analysis and Machine Intelligence. 32, 8 (2010), 1529--1535. DOI:https://doi.org/10.1109/TPAMI.2009.66.Google ScholarDigital Library
- Proença, H. and Alexandre, L.A. 2005. UBIRIS: A noisy iris image database. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (2005), 970--977.Google Scholar
- Kumar, A. and Passi, A. 2010. Comparison and combination of iris matchers for reliable personal authentication. Pattern Recognition. 43, 3 (2010), 1016--1026. DOI:https://doi.org/10.1016/j.patcog.2009.08.016.Google ScholarDigital Library
- Daugman, J. 2004. How Iris Recognition Works. IEEE Transactions on Circuits and Systems for Video Technology. 14, 1 (2004), 21--30. DOI:https://doi.org/10.1109/TCSVT.2003.818350.Google ScholarDigital Library
- Daugman, J. 2007. New methods in iris recognition. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics. 37, 5 (2007), 1167--1175. DOI:https://doi.org/10.1109/TSMCB.2007.903540.Google ScholarDigital Library
- He, Z. et al. 2009. Toward accurate and fast iris segmentation for iris biometrics. IEEE Transactions on Pattern Analysis and Machine Intelligence. 31, 9 (2009), 1670--1684. DOI:https://doi.org/10.1109/TPAMI.2008.183.Google ScholarDigital Library
- Chen, L.-C. et al. 2018. Rethinking Atrous Convolution for Semantic Image Segmentation Liang-Chieh. arXiv.org. (2018). DOI:https://doi.org/10.1159/000018039.Google Scholar
- Chen, L.C. et al. 2018. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence. 40, 4 (2018), 834--848. DOI:https://doi.org/10.1109/TPAMI.2017.2699184.Google ScholarCross Ref
- Chen, L.C. et al. 2018. Encoder-decoder with atrous separable convolution for semantic image segmentation. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (2018), 833--851.Google Scholar
- He, K. et al. 2016. Deep residual learning for image recognition. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2016).Google Scholar
- Phillips, P.J. et al. 2005. Overview of the face recognition grand challenge. Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005 (2005).Google Scholar
- Tan, C.W. and Kumar, A. 2013. Towards online iris and periocular recognition under relaxed imaging constraints. IEEE Transactions on Image Processing. 22, 10 (2013), 3751--3765. DOI:https://doi.org/10.1109/TIP.2013.2260165.Google ScholarDigital Library
- Zhao, Z. and Kumar, A. 2015. An accurate iris segmentation framework under relaxed imaging constraints using total variation model. Proceedings of the IEEE International Conference on Computer Vision (2015).Google Scholar
- Proença, H. 2010. Iris recognition: On the segmentation of degraded images acquired in the visible wavelength. IEEE Transactions on Pattern Analysis and Machine Intelligence. (2010). DOI:https://doi.org/10.1109/TPAMI.2009.140.Google ScholarDigital Library
- NICE.I Evaluation: http://nice1.di.ubi.pt/evaluation.htm.Google Scholar
Index Terms
- Accurate Iris Segmentation for at-a-distance Acquired Iris/Face Images under Less Constrained Environment
Recommendations
Real-time Iris Recognition System for Non-Ideal Iris Images
ICCAE 2019: Proceedings of the 2019 11th International Conference on Computer and Automation EngineeringBiometrics is popular nowadays because of its very useful security applications. There are different biometric technologies but iris recognition system was considered the most reliable since human irises are unique and cannot be forged easily. The study ...
Fast and accurate personal identification based on iris biometric
In this paper, we propose an efficient and accurate technique to process iris biometric. Our approach handles degraded quality of eye images, occlusion of eyelids and eyelashes. Our approach considers non-circular pupil and iris boundary. With our ...
Efficient iris segmentation method in unconstrained environments
Recently, iris recognition systems have gained increased attention especially in non-cooperative environments. One of the crucial steps in the iris recognition system is the iris segmentation because it significantly affects the accuracy of the feature ...
Comments