Using dynamic data augmentation and contrastive learning for iris verification

Abstract

Iris verification has gained extensive attention because of its uniqueness, stability, and non-invasiveness. Deep learning techniques have made significant progress in the field of iris verification. By using convolutional neural networks (CNNs), features of iris images can be automatically extracted and learned, enabling high-precision identity verification. However, challenges such as intra-class variability and limited dataset size can compromise verification accuracy. To address these issues, we proposed an iris verification method based on dynamic data augmentation and contrastive learning. Four data augmentation strategies were carefully designed for online iris enhancement and dataset expansion, and the accuracy of iris verification was further improved by using data augmentation probability scheduler(DAPS). The MobileNetV3 was employed as the backbone network, which was optimized with contrastive learning for 3-channel iris pairs. The proposed method was evaluated on two benchmark iris databases, CASIA-V4-Interval and CASIA-V4-Thousand, achieving high accuracies of 99.85% and 98.82%, respectively. Experimental results demonstrate that the proposed method can achieve competitive performance with a small number of training samples.

Keywords

iris verification / contrastive learning / convolutional neural network (CNN) / data augmentation

References

[1] BOWYER K W, HOLLINGSWORTH K, FLYNN P J. Image understanding for iris biometrics: A survey. Computer Vision and Image Understanding, 2008, 110(2): 281-307.

[2] LIU N, ZHANG M, LI H, et al. DeepIris: Learning pairwise filter bank for heterogeneous iris verification. Pattern Recognition Letters, 2016, 82: 154-61.

[3] De MARSICO M, PETROSINO A, RICCIARDI S. Iris recognition through machine learning techniques: A survey. Pattern Recognition Letters, 2016, 82: 106-15.

[4] ALINIA LAT R, DANISHVAR S, HERAVI H, et al. Boosting iris recognition by margin-based loss functions. Algorithms, 2022, 15(4): 118.

[5] MARAM G A, LAMIAA A E. Convolutional neural network based feature extraction for iris recognition. International Journal of Computer Science & Information Technology, 2018, 10: 65–78.

[6] ZHAO T, LIU Y, HUO G, et al. A deep learning iris recognition method based on capsule network architecture. IEEE Access, 2019, 7: 49691-49701.

[7] KONG A W K, ZHANG D, KAMEL M S. An analysis of iriscode. IEEE Transactions on Image Processing, 2010, 19(2): 522-532.

[8] KOCH G, ZEMEL R, SALAKHUTDINOV R. Siamese neural networks for one-shot image recognition//The 32nd International Conference on Machine Learning, July 6-11, 2015, Lille, France. New York: Curran Associates, Inc., 2015, 2: 1-8.

[9] HOWARD A, SANDLER M, CHU G, et al. Searching for mobilenetv3//IEEE/CVF Conference on Computer Vision,October 27-November 1, 2019, Seoul, South Korea. Piscataway, N.J.: IEEE, 2019: 1. arXiv:1905.02244

[10] DAUGMAN J. New methods in iris recognition. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics. 2007, 37(5): 1167-1175.

[11] DAUGMAN J. How iris recognition works. IEEE Transactions on Circuits & Systems for Video Technology, 2004, 14(1): 21-30.

[12] DAUGMAN J. Information theory and the iriscode. IEEE Transactions on Information Forensics and Security, 2015, 11(2): 400-409.

[13] DAUGMAN J G. High confidence visual recognition of persons by a test of statistical independence. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1993, 15(11): 1148-1161.

[14] YAO P, LI J, YE X, et al. Iris recognition algorithm using modified log-gabor filters//The 18th International Conference on Pattern Recognition, August 20-24, 2006 Hong Kong, China. Piscataway, N.J.: IEEE, 2006, 4: 461-464.

[15] SARHAN A M. Iris recognition using discrete cosine transform and artificial neural networks. Journal of Computer Science, 2009, 5(5): 369.

[16] SHI J X, GU X F. The comparison of iris recognition using principal component analysis, independent component analysis and Gabor wavelets//The 3rd International Conference on Computer Science and Information Technology, April 23-25, 2010. Piscataway, N.J.: IEEE. 2010: 5563947.

[17] JASIM Y A, AL-ANI A A, AL-ANI L A. Iris recognition using principal component analysis//The 1st Annual International Conference on Information and Sciences, November 20-21, 2018, University of Fallujah, Iraq. Piscataway, N.J.: IEEE, 2018: 00028.

[18] RASOOL R A. Iris feature extraction and recognition based on gray level co-occurrence matrix (GLCM) technique. International Journal of Computer Applications, 2018, 181(25): 15-17.

[19] HAJARI K, GAWANDE U, GOLHAR Y. Neural network approach to iris recognition in noisy environment. Procedia Computer Science, 2016, 78: 675-682.

[20] RASTI P, DANESHMAND M, ANBARJAFARI G. Statistical approach based iris recognition using local binary pattern. DYNA-Ingeniería e Industria, 2017, 92(1): 1.

[21] LOWE D G. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 2004, 60(2): 91-110.

[22] ALONSO-FERNANDEZ F, TOME-GONZALEZ P, RUIZ-ALBACETE V, et al. Iris recognition based on sift features//The 1st IEEE International Conference on Biometrics, Identity and Security, September 22-23, Tampa, USA. 2009. Piscataway, N.J.: IEEE, 2009: 1-8.

[23] SAHMOUD S A, ABUHAIBA I S. Efficient iris segmentation method in unconstrained environments. Pattern Recognition, 2013, 46(12): 3174-3185.

[24] REN X, PENG Z, ZENG Q, et al. An improved method for Daugman’s iris localization algorithm. Computers in Biology and Medicine, 2008, 38(1): 111-115.

[25] NGUYEN K, FOOKES C, ROSS A, et al. Iris recognition with off-the-shelf CNN features: A deep learning perspective. IEEE Access, 2018, 6: 18848-18855.

[26] GANGWAR A, JOSHI A. DeepIrisNet: Deep iris representation with applications in iris recognition and cross-sensor iris recognition//IEEE International Conference on Image Processing, September 25-28, 2016, Phoenix, Arizona, USA. 2016: 2301-2305.

[27] NGUYEN K, PROENCA H, ALONSO-FERNANDEZ F. Deep learning for iris recognition: A Survey. 2022. arXiv:221005866.

[28] LIU G, ZHOU W, TIAN L, et al. An efficient and accurate iris recognition algorithm based on a novel condensed 2-ch deep convolutional neural network. Sensors, 2021, 21(11): 3721.

[29] OTHMAN N, DORIZZI B, GARCIA-SALICETTI S. OSIRIS: An open source iris recognition software. Pattern Recognition Letters, 2016, 82: 124-131.

[30] SANPACHAI H, MALISUWAN S. A study of image enhancement for iris recognition. Journal of Industrial and Intelligent Information, 2015, 3(1): 61-64.

[31] CASIA-V4. [2023-08-29].Available online: http://biometrics.idealtest.org/

[32] PROENçA H, NEVES J C. Segmentation-less and non-holistic deep-learning frameworks for iris recognition//IEEE/CVF Conference on Computer Vision and Pattern Recognition, October 27-November 1, 2019, Seoul, South Korea. Piscataway, N.J.: IEEE, 2019: 1-10.

[33] NGUYEN K, FOOKES C, SRIDHARAN S. Constrained design of deep iris networks. IEEE Transactions on Image Processing, 2020, 29: 7166-7175.

[34] WEI J, HE R, SUN Z. Contrastive uncertainty learning for iris recognition with insufficient labeled samples//IEEE International Joint Conference on Biometrics, August 4-7, 2021, Shenzhen, China. Piscataway, N.J.: IEEE, 2021: 1-8.