TipSegNet: Fingertip Segmentation in Contactless Fingerprint Imaging
Abstract
:1. Introduction
1.1. Related Work
1.2. Contribution
- Novel Model Design creating TipSegNet: Utilizing transfer learning to create a ResNeXt-101-based feature extractor with a FPN-like decoder design for segmenting fingertips in hand images.
- Extended Data Augmentation: Augmenting the dataset with various transformations, such as perspective change, resizing and cropping, and solarization, thereby improving the model’s robustness to variations in contactless fingerprint recordings, specifically addressing changes like varying lighting conditions, different finger poses, and reducing overfitting in the training process.
- Comparison with SOTA: Comparing our model against established, traditional, and state-of-the-art methods, leading to the demonstration of superior segmentation performance in both cases.
2. Methods
2.1. Segmentation Using Deep Learning
2.1.1. Pre-Processing
- Resize + Crop: The image is randomly cropped to a region of a size of 0.75 to 1 times the original size, with an aspect ratio of between 0.9 and 1.1 of the original image, before it is padded to pixels.
- Rotation: The image is randomly rotated with an angle ranging from to 60 degrees.
- Perspective Change: This technique simulates random changes in the viewpoint by distorting the image accordingly.
- Gaussian Blur: A Gaussian blur is applied to the image, simulating various degrees of focus and sensor noise.
- Solarize: This technique inverts all pixel values above a certain threshold. It creates high-contrast images and simulates the ridge-valley inversion [34].
- Posterize: The number of bits used to represent the pixel values is reduced, decreasing the number of possible shades of gray in the image. This simplification simulates low-contrast recording, where the background is hard to separate from the fingertips.
- Histogram Equalization: This method adjusts the contrast of the image by spreading out the most frequent intensity values. It is a common enhancing technique used to improve the visual appearance and downstream performance of fingerprints.
2.1.2. Architecture
ResNeXt Family
FPN and Feature Hierarchy
Model Parameters and FLOPs
2.1.3. Training and Hyperparameter
2.2. Experiment
3. Results
3.1. Failure Case Analysis
3.2. Data Augmentation Ablation
3.3. Model Ablation
4. Discussion
- The use of ResNeXt-101 as a backbone: ResNeXt-101, with its concept of cardinality, captures a richer set of feature representations than traditional ResNet architectures. This is crucial for distinguishing subtle differences between fingertip regions and complex backgrounds.
- Feature Pyramid Network (FPN) integration: The FPN effectively combines multi-scale features, allowing the model to accurately segment fingertips regardless of their size or orientation in the image. This addresses a common challenge in contactless fingerprint imaging, where the finger pose can vary significantly.
- Extensive data augmentation: Our comprehensive augmentation strategy, including geometric transformations and intensity adjustments, significantly improves the model’s ability to generalize to diverse real-world scenarios. This is evident from the minimal difference in performance between the training and validation sets, indicating robustness to variations in image quality and capture conditions.
Limitations
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Yoon, S.; Jain, A.K. Longitudinal study of fingerprint recognition. Proc. Natl. Acad. Sci. USA 2015, 112, 8555–8560. [Google Scholar] [CrossRef] [PubMed]
- Birajadar, P.; Haria, M.; Kulkarni, P.; Gupta, S.; Joshi, P.; Singh, B.; Gadre, V. Towards smartphone-based touchless fingerprint recognition. Sādhanā 2019, 44, 161. [Google Scholar] [CrossRef]
- Kauba, C.; Söllinger, D.; Kirchgasser, S.; Weissenfeld, A.; Fernández Domínguez, G.; Strobl, B.; Uhl, A. Towards Using Police Officers’ Business Smartphones for Contactless Fingerprint Acquisition and Enabling Fingerprint Comparison against Contact-Based Datasets. Sensors 2021, 21, 2248. [Google Scholar] [CrossRef]
- Weissenfeld, A.; Schmid, R.; Kohn, B.; Strobl, B.; Domínguez, G.F. Case study of the acquisition of contactless fingerprints in a real police setting. In Proceedings of the 2022 International Conference of the Biometrics Special Interest Group (BIOSIG), Darmstadt, Germany, 14–16 September 2022; pp. 1–5468. [Google Scholar] [CrossRef]
- Ruzicka, L.; Söllinger, D.; Kohn, B.; Heitzinger, C.; Uhl, A.; Strobl, B. Improving Sensor Interoperability between Contactless and Contact-Based Fingerprints Using Pose Correction and Unwarping. IET Biom. 2023, 2023, 7519499. [Google Scholar] [CrossRef]
- Tan, H.; Kumar, A. Towards More Accurate Contactless Fingerprint Minutiae Extraction and Pose-Invariant Matching. IEEE Trans. Inf. Forensics Secur. 2020, 15, 3924–3937. [Google Scholar] [CrossRef]
- Grosz, S.A.; Engelsma, J.J.; Liu, E.; Jain, A.K. C2CL: Contact to Contactless Fingerprint Matching. arXiv 2021, arXiv:2104.02811. [Google Scholar] [CrossRef]
- Bazen, A.M.; Gerez, S.H. Segmentation of Fingerprint Images. In Proceedings of the ProRISC 2001 Workshop on Circuits, Systems and Signal Processing, Veldhoven, The Netherlands, 29–30 November 2001. [Google Scholar]
- Murshed, M.G.S.; Bahmani, K.; Schuckers, S.; Hussain, F. Deep Age-Invariant Fingerprint Segmentation System. arXiv 2023, arXiv:2303.03341. [Google Scholar] [CrossRef]
- Priesnitz, J.; Rathgeb, C.; Buchmann, N.; Busch, C. Deep Learning-Based Semantic Segmentation for Touchless Fingerprint Recognition. In Proceedings of the Pattern Recognition. ICPR International Workshops and Challenges, Virtual Event, 10–15 January 2021; Del Bimbo, A., Cucchiara, R., Sclaroff, S., Farinella, G.M., Mei, T., Bertini, M., Escalante, H.J., Vezzani, R., Eds.; Lecture Notes in Computer Science. Springer: Cham, Switzerland, 2021; pp. 154–168. [Google Scholar] [CrossRef]
- Zheng, X.; Wang, Y.; Zhao, X. Fingerprint Image Segmentation Using Active Contour Model. In Proceedings of the Fourth International Conference on Image and Graphics (ICIG 2007), Washington, DC, USA, 22–24 August 2007; pp. 437–441. [Google Scholar] [CrossRef]
- Murshed, M.G.S.; Abbas, S.K.; Purnapatra, S.; Hou, D.; Hussain, F. Deep Learning-Based Approaches for Contactless Fingerprints Segmentation and Extraction. arXiv 2023, arXiv:2311.15163. [Google Scholar] [CrossRef]
- Xie, S.; Girshick, R.; Dollar, P.; Tu, Z.; He, K. Aggregated Residual Transformations for Deep Neural Networks. In Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 21–26 July 2017; pp. 5987–5995. [Google Scholar] [CrossRef]
- Lin, T.Y.; Dollár, P.; Girshick, R.; He, K.; Hariharan, B.; Belongie, S. Feature Pyramid Networks for Object Detection. arXiv 2017, arXiv:1612.03144. [Google Scholar] [CrossRef]
- Garcia-Garcia, A.; Orts-Escolano, S.; Oprea, S.; Villena-Martinez, V.; Garcia-Rodriguez, J. A Review on Deep Learning Techniques Applied to Semantic Segmentation. arXiv 2017, arXiv:1704.06857. [Google Scholar] [CrossRef]
- Ghosh, S.; Das, N.; Das, I.; Maulik, U. Understanding Deep Learning Techniques for Image Segmentation. arXiv 2019, arXiv:1907.06119. [Google Scholar] [CrossRef]
- Sollinger, D.; Uhl, A. Optimizing contactless to contact-based fingerprint comparison using simple parametric warping models. In Proceedings of the 2021 IEEE International Joint Conference on Biometrics (IJCB), Shenzhen, China, 4–7 August 2021; pp. 1–7. [Google Scholar] [CrossRef]
- Ronneberger, O.; Fischer, P.; Brox, T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In Proceedings of the Medical Image Computing and Computer-Assisted Intervention–MICCAI, Munich, Germany, 5–9 October 2015; Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F., Eds.; Lecture Notes in Computer Science. Springer: Cham, Switzerland, 2015; pp. 234–241. [Google Scholar] [CrossRef]
- Tan, M.; Le, Q.V. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. arXiv 2020, arXiv:1905.11946. [Google Scholar] [CrossRef]
- Beheshti, N.; Johnsson, L. Squeeze U-Net: A Memory and Energy Efficient Image Segmentation Network. In Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Seattle, WA, USA, 14–19 June 2020; pp. 1495–1504. [Google Scholar] [CrossRef]
- He, K.; Gkioxari, G.; Dollár, P.; Girshick, R. Mask R-CNN. In Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 22–29 October 2017; pp. 2980–2988. [Google Scholar] [CrossRef]
- Chen, L.C.; Papandreou, G.; Kokkinos, I.; Murphy, K.; Yuille, A. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 40, 834–848. [Google Scholar] [CrossRef]
- Badrinarayanan, V.; Kendall, A.; Cipolla, R. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 39, 2481–2495. [Google Scholar] [CrossRef]
- Wang, J.; Sun, K.; Cheng, T.; Jiang, B.; Deng, C.; Zhao, Y.; Liu, D.; Mu, Y.; Tan, M.; Wang, X.; et al. Deep High-Resolution Representation Learning for Visual Recognition. IEEE Trans. Pattern Anal. Mach. Intell. 2021, 43, 3349–3364. [Google Scholar] [CrossRef]
- Priesnitz, J.; Huesmann, R.; Rathgeb, C.; Buchmann, N.; Busch, C. Mobile Contactless Fingerprint Recognition: Implementation, Performance and Usability Aspects. Sensors 2022, 22, 792. [Google Scholar] [CrossRef]
- Chen, Q.; Shen, F.; Ding, Y.; Gong, P.; Tao, Y.; Wang, J. Face Detection Using R-FCN Based Deformable Convolutional Networks. In Proceedings of the 2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Miyazaki, Japan, 7–10 October 2018; pp. 4165–4170. [Google Scholar] [CrossRef]
- Inácio, A.D.S.; Brilhador, A.; Lopes, H.S. Semantic Segmentation of Clothes in the Context of Soft Biometrics Using Deep Learning Methods. In Proceedings of the Anais do 14. Congresso Brasileiro de Inteligência Computacional. ABRICOM, Belèm, Brasil, 1 January 2020; pp. 1–7. Available online: https://sbic.org.br/eventos/cbic_2019/cbic2019-77/ (accessed on 8 September 2024).
- Lucio, D.R.; Laroca, R.; Zanlorensi, L.A.; Moreira, G.; Menotti, D. Simultaneous Iris and Periocular Region Detection Using Coarse Annotations. In Proceedings of the 2019 32nd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), Rio de Janeiro, Brazil, 28–30 October 2019; pp. 178–185. [Google Scholar] [CrossRef]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep Residual Learning for Image Recognition. arXiv 2015, arXiv:1512.03385. [Google Scholar] [CrossRef]
- Duc, T.L.; Anh, L.N.H.; Quoc, T.N.; Hoang, V.T. Ear Images Classification Based on Data Augmentation and ResNeXt50. In Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2022), Online, 14–16 December 2022; Abraham, A., Hanne, T., Gandhi, N., Manghirmalani Mishra, P., Bajaj, A., Siarry, P., Eds.; Springer Nature Switzerland: Cham, Switzerland, 2023; Volume 648, pp. 315–324. [Google Scholar] [CrossRef]
- Yang, Y.; Yang, Q.; Su, Z.; Wu, W. Employee Fingerprint Identification Management System Based on Bidirectional ResNext Network with Triplet Loss. In Proceedings of the 2024 8th International Conference on Management Engineering, Software Engineering and Service Sciences (ICMSS), Wuhan, China, 12–14 January 2024; pp. 48–52. [Google Scholar] [CrossRef]
- Labati, R.D.; Genovese, A.; Piuri, V.; Scotti, F. Contactless fingerprint recognition: A neural approach for perspective and rotation effects reduction. In Proceedings of the 2013 IEEE Symposium on Computational Intelligence in Biometrics and Identity Management (CIBIM), Singapore, 16–19 April 2013; pp. 22–30, ISSN 325-4319. [Google Scholar] [CrossRef]
- Chowdhury, A.M.; Imtiaz, M. Contactless Fingerprint Recognition Using Deep Learning—A Systematic Review. J. Cybersecur. Priv. 2022, 2, 714–730. [Google Scholar] [CrossRef]
- Vyas, R.; Kumar, A. A Collaborative Approach using Ridge-Valley Minutiae for More Accurate Contactless Fingerprint Identification. arXiv 2019, arXiv:1909.06045. [Google Scholar]
- Mahajan, D.; Girshick, R.; Ramanathan, V.; He, K.; Paluri, M.; Li, Y.; Bharambe, A.; van der Maaten, L. Exploring the Limits of Weakly Supervised Pretraining. arXiv 2018, arXiv:1805.00932. [Google Scholar]
- Iakubovskii, P. Segmentation Models Pytorch. 2019. Available online: https://segmentation-modelspytorch.readthedocs.io/en/latest/ (accessed on 14 June 2024).
- Zhai, X.; Kolesnikov, A.; Houlsby, N.; Beyer, L. Scaling Vision Transformers. arXiv 2022, arXiv:2106.04560. [Google Scholar] [CrossRef]
- Wang, Z.; Ning, X.; Blaschko, M.B. Jaccard Metric Losses: Optimizing the Jaccard Index with Soft Labels. arXiv 2024, arXiv:2302.05666. [Google Scholar]
- Lee, C.; Lee, S.; Kim, J.; Kim, S.J. Preprocessing of a Fingerprint Image Captured with a Mobile Camera. In Advances in Biometrics; Zhang, D., Jain, A.K., Eds.; Springer: Berlin/Heidelberg, Germany, 2005; pp. 348–355. [Google Scholar] [CrossRef]
- Raghavendra, R.; Busch, C.; Yang, B. Scaling-robust fingerprint verification with smartphone camera in real-life scenarios. In Proceedings of the 2013 IEEE Sixth International Conference on Biometrics: Theory, Applications and Systems (BTAS), Arlington, VA, USA, 29 September–2 October 2013; pp. 1–8. [Google Scholar] [CrossRef]
- Chetlur, S.; Woolley, C.; Vandermersch, P.; Cohen, J.; Tran, J.; Catanzaro, B.; Shelhamer, E. cuDNN: Efficient Primitives for Deep Learning. arXiv 2014, arXiv:1410.0759. [Google Scholar] [CrossRef]
- Hinton, G.; Vinyals, O.; Dean, J. Distilling the Knowledge in a Neural Network. arXiv 2015, arXiv:1503.02531. [Google Scholar] [CrossRef]
- Wang, L.; Yoon, K.J. Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New Outlooks. IEEE Trans. Pattern Anal. Mach. Intell. 2022, 44, 3048–3068. [Google Scholar] [CrossRef]
- Hong, L.; Wan, Y.; Jain, A. Fingerprint image enhancement: Algorithm and performance evaluation. IEEE Trans. Pattern Anal. Mach. Intell. 1998, 20, 777–789. [Google Scholar] [CrossRef]
Model Part | Parameters | FLOPs |
---|---|---|
Encoder | ||
Decoder | ||
Segmentation Head | 1161 | |
Total |
Model | Parameters | FLOPs |
---|---|---|
ResNet-34 | ||
Encoder | ||
Decoder | ||
ResNet-50 | ||
Encoder | ||
Decoder | ||
ResNet-101 | ||
Encoder | ||
Decoder |
Approach | mIoU | Accuracy | |
---|---|---|---|
Group 1 | Otsu [10] 1 | 0.92 | - |
Color Histogram [3] 1 | 0.38 | - | |
Gaussian Mixture [3] 1 | 0.31 | - | |
Mask R-CNN [3] | 0.96 | - | |
DeepLabv3+ [10] | 0.95 | - | |
Group 2 | Segnet [3] | 0.90 | - |
HRNet [3] | 0.85 | - | |
DeepLab [3] | 0.93 | - | |
TipSegNet | 0.99 | 1.00 | |
Group 3 | Color & Texture [39] 2 | - | 0.99 |
Mean Shift [40] 2 | - | 0.92–0.96 | |
U-Net [5] 2 | 0.91 | 0.98 | |
EfficientNet [5] 2 | 0.50 | 0.88 | |
SqueezeNet [5] 2 | 0.86 | 0.96 |
Accuracy | F1 | IoU | Recall | |
---|---|---|---|---|
no augmentation | 0.999 | 0.994 | 0.987 | 0.994 |
minimal augmentation | 0.998 | 0.993 | 0.985 | 0.993 |
data augmentation | 0.999 | 0.994 | 0.987 | 0.994 |
Model | Accuracy | F1 | IoU | Recall |
---|---|---|---|---|
ResNet-34 | 0.998 | 0.990 | 0.981 | 0.990 |
ResNet-50 | 0.997 | 0.988 | 0.976 | 0.988 |
ResNet-101 | 0.998 | 0.992 | 0.984 | 0.992 |
ResNeXt-101 | 0.999 | 0.994 | 0.987 | 0.994 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ruzicka, L.; Kohn, B.; Heitzinger, C. TipSegNet: Fingertip Segmentation in Contactless Fingerprint Imaging. Sensors 2025, 25, 1824. https://doi.org/10.3390/s25061824
Ruzicka L, Kohn B, Heitzinger C. TipSegNet: Fingertip Segmentation in Contactless Fingerprint Imaging. Sensors. 2025; 25(6):1824. https://doi.org/10.3390/s25061824
Chicago/Turabian StyleRuzicka, Laurenz, Bernhard Kohn, and Clemens Heitzinger. 2025. "TipSegNet: Fingertip Segmentation in Contactless Fingerprint Imaging" Sensors 25, no. 6: 1824. https://doi.org/10.3390/s25061824
APA StyleRuzicka, L., Kohn, B., & Heitzinger, C. (2025). TipSegNet: Fingertip Segmentation in Contactless Fingerprint Imaging. Sensors, 25(6), 1824. https://doi.org/10.3390/s25061824