Evaluating the Impact of Iris Image Preprocessing on CNN-Based Biometric Recognition Performance
Article Sidebar
Main Article Content
Abstract
Iris-based biometric recognition is a reliable identification technique because the iris pattern is unique and remains stable throughout an individual's lifetime. However, the quality of iris images significantly affects the accuracy of recognition systems, particularly when images are degraded by noise, uneven illumination, and occlusions caused by eyelashes and eyelids. This study investigates the impact of several image preprocessing techniques on the performance of an iris biometric recognition system. The preprocessing pipeline consists of image resizing, pixel value normalization, contrast enhancement using Contrast Limited Adaptive Histogram Equalization (CLAHE), and noise reduction through Gaussian filtering. The CASIA-IrisV1 dataset was employed for model development and evaluation, while a Convolutional Neural Network (CNN) was used as the classification model. Experimental results demonstrate that the proposed preprocessing pipeline improves the iris recognition accuracy from 91.2% to 96.4%. These findings indicate that image quality enhancement through appropriate preprocessing plays a crucial role in improving the robustness and accuracy of CNN-based iris biometric recognition systems.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
References
[1] J. R. Malgheet, N. B. Manshor, and L. S. Affendey, “Iris recognition development techniques: A comprehensive review,” *Complexity*, vol. 2021, Art. no. 6641247, 32 pp., 2021, doi: 10.1155/2021/6641247.
[2] M. Jenadeleh, M. Pedersen, and D. Saupe, “Blind quality assessment of iris images acquired in visible light for biometric recognition,” *Sensors*, vol. 20, no. 5, Art. no. 1308, 2020, doi: 10.3390/s20051308.
[3] L. Wang, K. Zhang, M. Ren, Y. Wang, and Z. Sun, “Recognition oriented iris image quality assessment in the feature space,” in *Proc. IEEE Int. Joint Conf. Biometrics (IJCB)*, Houston, TX, USA, 2020, pp. 1–8, doi: 10.1109/IJCB48548.2020.9304896.
[4] A. Khan, A. Sohail, U. Zahoora, and A. S. Qureshi, “A survey of the recent architectures of deep convolutional neural networks,” *Artificial Intelligence Review*, vol. 53, no. 8, pp. 5455–5516, 2020, doi: 10.1007/s10462-020-09825-6.
[5] M. Trokielewicz, A. Czajka, and P. Maciejewicz, “Post-mortem iris recognition with deep-learning-based image segmentation,” *Image and Vision Computing*, vol. 94, Art. no. 103866, 2020, doi: 10.1016/j.imavis.2019.103866.
[6] J. R. Malgheet, N. B. Manshor, and L. S. Affendey, “Iris recognition development techniques: A comprehensive review,” *Complexity*, vol. 2021, Art. no. 6641247, 32 pp., 2021, doi: 10.1155/2021/6641247.
[7] B. Moktari *et al*., “Iris recognition system using convolutional neural network,” *International Journal of Computing and Digital Systems*, vol. 10, no. 1, 2021, doi: 10.12785/ijcds/100101.
[8] A. Agarwal, A. Noore, M. Vatsa, and R. Singh, “Enhanced iris presentation attack detection via contraction-expansion CNN,” *Pattern Recognition Letters*, vol. 159, pp. 61–69, 2022, doi: 10.1016/j.patrec.2022.04.007.
[9] A. Agarwal, A. Noore, M. Vatsa, and R. Singh, “Generalized contact lens iris presentation attack detection,” *IEEE Transactions on Biometrics, Behavior, and Identity Science*, vol. 4, no. 3, pp. 373–385, 2022, doi: 10.1109/TBIOM.2022.3177669.
[10] Y. Zhang, J. Muhammad, Y. Wang, K. Zhang, and Z. Sun, “Attention meta-transfer learning approach for few-shot iris recognition,” *Computers & Electrical Engineering*, vol. 99, Art. no. 107848, 2022, doi: 10.1016/j.compeleceng.2022.107848.
[11] H. Proença, “DeepGabor: A learning-based framework to augment IrisCodes permanence,” *IEEE Transactions on Information Forensics and Security*, vol. 17, pp. 3748–3757, 2022, doi: 10.1109/TIFS.2022.3214098.
[12] K. Nguyen, C. Fookes, S. Sridharan, and A. Ross, “Complex-valued iris recognition network,” *IEEE Transactions on Pattern Analysis and Machine Intelligence*, vol. 45, no. 1, pp. 182–196, 2023, doi: 10.1109/TPAMI.2022.3152857.
[13] Y. Yin, S. He, R. Zhang, H. Chang, X. Han, and J. Zhang, “Deep learning for iris recognition: A review,” *Neural Computing and Applications*, vol. 37, no. 17, pp. 11125–11173, 2025, doi: 10.1007/s00521-025-11109-5.
[14] K. Nguyen, H. Proença, and F. Alonso-Fernandez, “Deep learning for iris recognition: A survey,” *ACM Computing Surveys*, vol. 56, no. 9, Art. no. 223, pp. 1–35, 2024, doi: 10.1145/3651306.
[15] H. H. Rasheed, S. S. Shamini, M. A. Mahmoud, and M. A. Alomari, “Review of iris segmentation and recognition using deep learning to improve biometric application,” *Journal of Intelligent Systems*, vol. 32, no. 1, Art. no. 20230139, 2023, doi: 10.1515/jisys-2023-0139.
[16] J. Li, J. Muhammad, Y. Wang, K. Zhang, and Z. Sun, “Smartphone-based iris recognition through high-quality visible-spectrum iris image capture,” *Pattern Recognition*, vol. 138, Art. no. 109471, 2023, doi: 10.1016/j.patcog.2023.109471.
[17] J. E. Zambrano, J. I. Pilataxi, C. A. Perez, and K. W. Bowyer, “Iris recognition using an enhanced pre-trained backbone based on anti-aliased CNNs,” *IEEE Access*, vol. 12, pp. 94570–94583, 2024, doi: 10.1109/ACCESS.2024.3425648.
[18] N. Fakhraei, R. Donida Labati, V. Piuri, and F. Scotti, “Deep learning-based iris quality assessment for images sourced from websites and social media,” in *Proc. IEEE Int. Conf. Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)*, Piraeus, Greece, 2025, pp. 1–7, doi: 10.1109/CIVEMSA65862.2025.11084826.