Size Optimized Generative Adversarial Networks for Single Image Super-Resolution of Fingerprint Images for Big Data Applications
Keywords:
Fingerprint image, Generative adversarial network, Single image super-resolution, SIFT.Abstract
Big data applications such as the Aadhar project necessitate the storage of large volumes of biometric data, which requires about 20,218 TB of storage space. Big data projects, which present enormous storage and processing challenges, demand large storage servers and high-end computers. This paper suggests a novel method to enhance the resolution of compressed fingerprint images using a super-resolution (SR) model that reconstructs a high-resolution (HR) image from a low-resolution (LR) image can significantly reduce the requirement for large amounts of data and costly hardware in such cases. This proposed size optimized GAN (Generative Adversarial Network) based SR model of size 2.9 MB, that enlarges a low-resolution image to a scale factor of eight times. The model was trained on the fingerprint data set FVC 2004 (Fingerprint Verification Competition) and then tested on the FVC 2004 data set. The various human visual system (HVS) parameters, such as peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and mean squared error (MSE), were measured, and their values were found to be 35.97, 0.958, and 19.5, respectively. Perceptual loss was measured in terms of generator loss as 0.981 and discriminator loss as 0.550. The accuracy of matching between the ground-truth image and the regenerated image was measured through the SIFT (scale-invariant feature transform) method and obtained an identification accuracy of 98.7%. This approach has the potential to increase the performance of fingerprint recognition systems for latent fingerprint images.
Downloads
References
C. Applications, “Big Data Compression for Aadhaar Storage through Reduced Bit Level Ordering,” vol. 7, no. 2, pp. 59–64, 2020.
U. U. Deshpande, V. S. Malemath, S. M. Patil, and S. V. Chaugule, “CNNAI: A Convolution Neural Network-Based Latent Fingerprint Matching Using the Combination of Nearest Neighbor Arrangement Indexing,” Front. Robot. AI, vol. 7, no. September, 2020, doi: 10.3389/frobt.2020.00113.
A. Muhammed and A. R. Pais, “A Novel Fingerprint Image Enhancement based on Super Resolution,” 2020 6th Int. Conf. Adv. Comput. Commun. Syst. ICACCS 2020, no. March, pp. 165–170, 2020, doi: 10.1109/ICACCS48705.2020.9074196.
D. L. Andreea-Monica, S. Moldovanu, and L. Moraru, “A Fingerprint Matching Algorithm Using the Combination of Edge Features and Convolution Neural Networks,” Inventions, vol. 7, no. 2, pp. 1–14, 2022, doi: 10.3390/inventions7020039.
S. Saponara, A. Elhanashi, and Q. Zheng, “Recreating fingerprint images by convolutional neural network autoencoder architecture,” IEEE Access, vol. 9, pp. 147888–147899, 2021, doi: 10.1109/ACCESS.2021.3124746.
M. A. N. I. Fahim and H. Y. Jung, “A Lightweight GAN Network for Large Scale Fingerprint Generation,” IEEE Access, vol. 8, no. January, pp. 92918–92928, 2020, doi: 10.1109/ACCESS.2020.2994371.
I. Joshi, A. Anand, S. Dutta Roy, and P. Kumar Kalra, “On Training Generative Adversarial Network for Enhancement of Latent Fingerprints,” AI Deep Learn. Biometric Secur., vol. 1, pp. 51–79, 2021, doi: 10.1201/9781003003489-3.
K. Prajapati et al., “Unsupervised Single Image Super-Resolution Using Cycle Generative Adversarial Network,” Commun. Comput. Inf. Sci., vol. 1382, pp. 359–370, 2021, doi: 10.1007/978-3-030-71711-7_30.
S. Chatterjee, D. Hazra, Y. C. Byun, and Y. W. Kim, “Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation,” Mathematics, vol. 10, no. 9, pp. 1–16, 2022, doi: 10.3390/math10091541.
S. N. Ferdous, A. Dabouei, J. Dawson, and N. M. Nasrabadi, “Super-resolution guided pore detection for fingerprint recognition,” Proc. - Int. Conf. Pattern Recognit., pp. 2025–2032, 2020, doi: 10.1109/ICPR48806.2021.9413043.
Y. Zhu, X. Yin, and J. Hu, “FingerGAN: A Constrained Fingerprint Generation Scheme for Latent Fingerprint Enhancement,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 7, pp. 8358–8371, 2023, doi: 10.1109/TPAMI.2023.3236876.
Z. Yang, Y. Xu, and G. Lu, “Efficient Method for High-Resolution Fingerprint Image Enhancement Using Deep Residual Network,” 2020 IEEE Symp. Ser. Comput. Intell. SSCI 2020, pp. 1725–1730, 2020, doi: 10.1109/SSCI47803.2020.9308442.
S. Li, M. Zhao, Z. Fang, Y. Zhang, and H. Li, “Image Super-Resolution Using Lightweight Multiscale Residual Dense Network,” Int. J. Opt., vol. 2020, 2020, doi: 10.1155/2020/2852865.
H. Shams, T. Jan, A. A. Khalil, N. Ahmad, A. Munir, and R. A. Khalil, “Fingerprint image enhancement using multiple filters,” PeerJ Comput. Sci., vol. 9, pp. 1–19, 2023, doi: 10.7717/peerj-cs.1183.
Y. Li, N. Xiao, and W. Ouyang, “Improved boundary equilibrium generative adversarial networks,” IEEE Access, vol. 6, pp. 11342–11348, 2018, doi: 10.1109/ACCESS.2018.2804278.
A. Radford, L. Metz, and S. Chintala, “Unsupervised representation learning with deep convolutional generative adversarial networks,” 4th Int. Conf. Learn. Represent. ICLR 2016 - Conf. Track Proc., pp. 1–16, 2016.
M. Arjovsky, S. Chintala, and L. Bottou, “Wasserstein GAN,” 2017, [Online]. Available: http://arxiv.org/abs/1701.07875
P. Shamsolmoali, M. Zareapoor, R. Wang, D. K. Jain, and J. Yang, “G-GANISR: Gradual generative adversarial network for image super resolution,” Neurocomputing, vol. 366, pp. 140–153, 2019, doi: 10.1016/j.neucom.2019.07.094.
A. Kumar, G. Ghinea, and S. Merugu, “Proceedings of the 2nd International Conference on Cognitive and Intelligent Computing,” vol. 2, no. March, pp. 27–28, 2022, doi: 10.1007/978-981-19-2358-6.
R. Gupta, M. Khari, D. Gupta, and R. G. Crespo, “Fingerprint image enhancement and reconstruction using the orientation and phase reconstruction,” Inf. Sci. (Ny)., vol. 530, pp. 201–218, 2020, doi: 10.1016/j.ins.2020.01.031.
S. V. de S. Neto, L. V. Batista, P. I. A. Guimarães, and T. E. S. de Souza, “Fingerprint image enhancement using fully convolutional deep autoencoders / Destaque de imagens de impressão digital utilizando autoencoders profundos totalmente convolucionais,” Brazilian J. Dev., vol. 8, no. 5, pp. 40027–40042, 2022, doi: 10.34117/bjdv8n5-474.
D. Maio, D. Maltoni, R. Cappelli, J. L. Wayman, and A. K. Jain, “FVC2004: Third fingerprint verification competition,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 3072, no. January, pp. 1–7, 2004, doi: 10.1007/978-3-540-25948-0_1.
C. S. Varnan, A. Jagan, J. Kaur, D. Jyoti, and D. S. Rao, “Image Quality Assessment Techniques in Spatial Domain,” Int. J. Comput. Sci. Technol., vol. 2, no. 3, pp. 177–184, 2011.
K. A. Yusharyahya, A. S. Nugroho, J. Purnama, and M. Galsinium, “A comparison of fingerprint enhancement algorithms for poor quality fingerprint images,” Proc. - 2014 Int. Conf. Adv. Informatics Concept, Theory Appl. ICAICTA 2014, pp. 342–347, 2015, doi: 10.1109/ICAICTA.2014.7005966.
M. Akif and B. I. N. Fauzi, “Fingerprint Recognition Using Scale Invariant Feature Transform Algorithm,” vol. 5, no. 04, pp. 719–721, 2017.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
