Reinforced Palmprint Reconstruction Attacks in Biometric Systems
Abstract
:1. Introduction
- (1)
- In order to comprehensively evaluate biometric reconstruction attacks in identity management systems, more indicators are proposed, including similarity, naturalness, visual quality, and completeness.
- (2)
- Any palmprint, which can be easily obtained, is used as the initial image, so naturalness can be ensured. The ROI of the initial image is iteratively modified with deep reinforcement strategies to reduce the matching distance. There is no remarkable sudden change near the boundaries of ROI in the complete fake image, so both visual quality and completeness can be satisfied.
- (3)
- In the first attack, Modification Constraint within Neighborhood (MCwN) is proposed to limit the modification extent and suppress the reckless modification to enhance the naturalness and visual quality.
- (4)
- In the second attack, Batch Member Selection (BMS) is proposed to select the significant pixels (SPs) to compose the batch, in which the SPs, i.e., the batch members, are modified simultaneously to reduce the matching number, i.e., computational complexity. Since the pixels in the batch are modified together, their modifications are slighter, and accordingly both the naturalness and visual quality are maximized.
2. Related Works
2.1. Reconstruction Attacks
2.2. Palmprint Recognition
3. Methodology
3.1. Modification Constraint within Neighborhood
3.2. Batch Member Selection
4. Experiments
4.1. Dataset and Palmprint Recognition Methods
4.2. Attack Performance
4.3. Ablation Experiment
5. Conclusions and Future Works
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Yang, W.; Wang, S.; Sahri, N.M.; Karie, N.M.; Ahmed, M.; Valli, C. Biometrics for Internet-of-Things Security: A Review. Sensors 2021, 21, 6163. [Google Scholar] [CrossRef] [PubMed]
- Nagar, A.; Nandakumar, K.; Jain, A.K. Biometric template transformation: A security analysis. In Proceedings of the SPIE, Electronic Imaging, Media Forensics and Security XII, San Jose, CA, USA, 17–21 January 2010; p. 75410. [Google Scholar]
- Gomez-Barrero, M.; Galbally, J. Reversing the irreversible: A survey on inverse biometrics. Comput. Secur. 2020, 90, 101700. [Google Scholar] [CrossRef]
- Ali, A.; Hoque, S.; Deravi, F. Directed Gaze Trajectories for biometric presentation attack detection. Sensors 2021, 21, 1394. [Google Scholar] [CrossRef] [PubMed]
- M Jomaa, R.; Mathkour, H.; Bazi, Y.; Islam, M.S. End-to-End Deep Learning Fusion of Fingerprint and Electrocardiogram Signals for Presentation Attack Detection. Sensors 2020, 20, 2085. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Hill, C.J. Risk of Masquerade Arising from the Storage of Biometrics. Bachelor’s Thesis, The Department of Computer Science, Australian National University, Canberra, Australia, 2001. [Google Scholar]
- Uludag, U.; Jain, A.K. Attacks on biometric systems: A case study in fingerprints. In Proceedings of the SPIE-EI, Security, Seganography and Watermarking of Multimedia Contents VI, San Jose, CA, USA, 16–20 June 2004; pp. 622–633. [Google Scholar]
- Ross, A.; Shah, J.; Jain, A.K. From template to image: Reconstructing fingerprints from minutiae points. IEEE Trans. Pattern Anal. Mach. Intell. 2007, 29, 544–560. [Google Scholar] [CrossRef]
- Cappelli, R.; Maio, D.; Lumini, A.; Maltoni, D. Fingerprint Image Reconstruction from Standard Templates. IEEE Trans. Pattern Anal. Mach. Intell. 2007, 29, 1489–1503. [Google Scholar] [CrossRef] [PubMed]
- Feng, J.; Jain, A.K. FM model based fingerprint reconstruction from minutiae template. In Proceedings of the International Conference on Biometrics, Alghero, Italy, 2–5 June 2009; pp. 544–553. [Google Scholar]
- Feng, J.; Jain, A.K. Fingerprint reconstruction: From minutiae to phase. IEEE Trans. Pattern Anal. Mach. Intell. 2011, 33, 209–223. [Google Scholar] [CrossRef] [Green Version]
- Li, S.; Kot, A.C. An improved scheme for full fingerprint reconstruction. IEEE Trans. Inf. Forensics Secur. 2012, 7, 1906–1912. [Google Scholar] [CrossRef]
- Cao, K.; Jain, A.K. Learning fingerprint reconstruction: From minutiae to image. IEEE Trans. Inf. Forensics Secur. 2015, 10, 104–117. [Google Scholar] [CrossRef] [Green Version]
- Kim, H.; Cui, X.; Kim, M.; Nguyen, T.H.B. Reconstruction of fingerprints from minutiae using conditional adversarial networks. Int. Workshop Digit. Watermarking 2018, 11378, 353–362. [Google Scholar]
- Adler, A. Sample images can be independently restored from face recognition templates. In Proceedings of the Canadian Conference on Electrical and Computer Engineering, Montreal, QC, Canada, 4–7 May 2003; pp. 1163–1166. [Google Scholar]
- Adler, A. Images can be regenerated from quantized biometric match score data. In Proceedings of the Canadian Conference on Electrical and Computer Engineering, Niagara Falls, ON, Canada, 2–5 May 2004; pp. 469–472. [Google Scholar]
- Mohanty, P.; Sarkar, S.; Kasturi, R. From scores to face templates: A model-based approach. IEEE Trans. Pattern Anal. Mach. Intell. 2007, 29, 2065–2078. [Google Scholar] [CrossRef]
- Galbally, J.; Fierrez, J.; Ortega-Garcia, J.; McCool, C.; Marcel, S. Hill-climbing attack to an eigenface-based face verification system. In Proceedings of the 2009 First IEEE International Conference on Biometrics, Identity and Security (BIdS), Tampa, FL, USA, 22–23 September 2009; pp. 1–6. [Google Scholar]
- Galbally, J.; McCool, C.; Fierrez, J.; Marcel, S.; Ortega-Garcia, J. On the vulnerability of face verification systems to hill-climbing attacks. Pattern Recognit. 2010, 43, 1027–1038. [Google Scholar] [CrossRef]
- Gomez-Barrero, M.; Galbally, J.; Fierrez, J.; Ortega-Garcia, J. Face verification put to test: A hill-climbing attack based on the uphill-simplex algorithm. In Proceedings of the 2012 5th IAPR International Conference on Biometrics (ICB), New Delhi, India, 29 March–1 April 2012; pp. 40–45. [Google Scholar]
- Mignon, A.; Jurie, F. Reconstructing faces from their signatures using RBF regression. In Proceedings of the British Machine Vision Conference 2013, Bristol, UK, 9–13 September 2013; pp. 103.1–103.12. [Google Scholar]
- Feng, Y.C.; Lim, M.H.; Yuen, P.C. Masquerade attack on transform-based binary-template protection based on perceptron learning. Pattern Recognit. 2014, 47, 3019–3033. [Google Scholar] [CrossRef]
- Mai, G.; Cao, K.; Yuen, P.C.; Jain, A.K. On the reconstruction of face images from deep face templates. IEEE Trans. Pattern Anal. Mach. Intell. 2018, 41, 1188–1202. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Rathgeb, C.; Uhl, A. Attacking iris recognition: An efficient hill-climbing technique. In Proceedings of the IEEE/IAPR International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, 23–26 August 2010; pp. 1217–1220. [Google Scholar]
- Venugopalan, S.; Savvides, M. How to generate spoofed irises from an iris code template. IEEE Trans. Inf. Forensics Secur. 2011, 6, 385–395. [Google Scholar] [CrossRef]
- Galbally, J.; Ross, A.; Gomez-Barrero, M.; Fierrez, J.; Ortega-Garcia, J. Iris image reconstruction from binary templates: An efficient probabilistic approach based on genetic algorithms. Comput. Vis. Image Underst. 2013, 117, 1512–1525. [Google Scholar] [CrossRef] [Green Version]
- Wang, F.; Leng, L.; Teoh, A.B.J.; Chu, J. Palmprint False Acceptance Attack with a Generative Adversarial Network (GAN). Appl. Sci. 2020, 10, 8547. [Google Scholar] [CrossRef]
- Bai, X.; Gao, N.; Zhang, Z.; Zhang, D. 3D palmprint identification combining blocked ST and PCA. Pattern Recognit. Lett. 2017, 100, 89–95. [Google Scholar] [CrossRef]
- Leng, L.; Zhang, J.; Khan, M.K.; Chen, X.; Alghathbar, K. Dynamic weighted discrimination power analysis: A novel approach for face and palmprint recognition in DCT domain. Int. J. Phys. Sci. 2010, 5, 2543–2554. [Google Scholar]
- Leng, L.; Li, M.; Kim, C.; Bi, X. Dual-source discrimination power analysis for multi-instance contactless palmprint recognition. Multimed. Tools Appl. 2017, 76, 333–354. [Google Scholar] [CrossRef]
- Fei, L.; Wen, J.; Zhang, Z.; Yan, K.; Zhong, Z. Local multiple directional pattern of palmprint image. In Proceedings of the 23rd International Conference on Pattern Recognition (ICPR), Cancun, Mexico, 4–8 December 2016; pp. 3013–3018. [Google Scholar]
- Liu, Y.; Kumar, A. Contactless palmprint identification using deeply learned residual features. IEEE Trans. Biom. Behav. Identity Sci. 2020, 2, 172–181. [Google Scholar] [CrossRef]
- Leng, L.; Yang, Z.; Min, W. Democratic voting downsampling for coding-based palmprint recognition. IET Biometr. 2020, 9, 290–296. [Google Scholar] [CrossRef]
- Fei, L.; Lu, G.; Jia, W.; Teng, S.; Zhang, D. Feature extraction methods for palmprint recognition: A survey and evaluation. IEEE Trans. Syst. Man Cybern. Syst. 2018, 49, 346–363. [Google Scholar] [CrossRef]
- Yang, Z.; Leng, L.; Min, W. Extreme downsampling and joint feature for coding-based palmprint recognition. IEEE Trans. Instrum. Meas. 2021, 70, 1–12. [Google Scholar] [CrossRef]
- Zhang, D.; Kong, W.; You, J.; Wong, M. Online palmprint identification. IEEE Trans. Pattern Anal. Mach. Intell. 2003, 25, 1041–1050. [Google Scholar] [CrossRef] [Green Version]
- Guo, Z.H.; Zhang, D.; Zhang, L.; Zuo, W.M. Palmprint verification using binary orientation co-occurrence vector. Pattern Recognit. Lett. 2009, 30, 1219–1227. [Google Scholar] [CrossRef]
- Sun, Z.; Tan, T.; Wang, Y.; Li, S.Z. Ordinal palmprint represention for personal identification. In Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), Washington, DC, USA, 20–25 June 2005; pp. 279–284. [Google Scholar]
- Kong, A.; Zhang, D.; Karmel, M. Palmprint identification using feature-level fusion. Pattern Recognit. 2006, 39, 478–487. [Google Scholar] [CrossRef]
- Kong, A.; Zhang, D. Competitive coding scheme for palmprint verification. In Proceedings of the International Conference on Pattern Recognition (ICPR), Cambridge, UK, 26–26 August 2004; pp. 520–523. [Google Scholar]
- Jia, W.; Huang, D.; Zhang, D. Palmprint verification based on robust line orientation code. Pattern Recognit. 2008, 41, 1504–1513. [Google Scholar] [CrossRef]
- Fei, L.; Xu, Y.; Tang, W.; Zhang, D. Double-orientation code and nonlinear matching scheme for palmprint recognition. Pattern Recognit. 2016, 49, 89–101. [Google Scholar] [CrossRef]
- Xu, Y.; Fei, L.; Wen, J.; Zhang, D. Discriminative and robust competitive code for palmprint recognition. IEEE Trans. Syst. Man Cybern. Syst. 2018, 48, 232–241. [Google Scholar] [CrossRef]
Ref. | Year | Modality | Methodology | Naturalness | Visual Quality |
---|---|---|---|---|---|
[6] | 2001 | Fingerprint | The orientations were reconstructed from the singular points (core, delta) based on pole zero model. Some lines were drawn through the details, resulting in only a sketch of the fingerprints. | L | M |
[7] | 2004 | Fingerprint | The minutiae image was reconstructed using HC. | L | L |
[8] | 2007 | Fingerprint | The direction, category and ridge of the original fingerprint were extracted from the minutiae template. | L | M |
[9] | 2007 | Fingerprint | Local detail model was used to initialize the image, and then Gabor filter was iteratively applied to the image formed by the detail parts. | M | M |
[10] | 2009 | Fingerprint | The orientation field was used to reconstruct the continuous phase that was combined with spiral phase. | M | M |
[11] | 2011 | Fingerprint | The phase image was reconstructed from fingerprint minutiae template, and then converted into a gray image. | M | H |
[12] | 2012 | Fingerprint | A binary ridge pattern was generated, which has a similar ridge flow to that of the original fingerprint. The continuous phase was intuitively reconstructed by removing the spirals in the phase image estimated from the ridge pattern. | M | H |
[13] | 2015 | Fingerprint | The prior knowledge of fingerprint ridge structure was coded through the direction patch and continuous phase patch dictionary. Then the direction field and ridge pattern were reconstructed. | M | H |
[14] | 2018 | Fingerprint | Fingerprint images were reconstructed using cGNA and fingerprint minutiae templates. | M | H |
[15] | 2003 | Face | A candidate image was slightly modified by an eigenface image, and the modifications improving the match score were kept. | L | L |
[16] | 2004 | Face | Face images were reconstructed using HC. | L | L |
[17] | 2007 | Face | Given the coordinates of the targeted subject in the affine space, the original template was reconstructed based on inverse affine transformation. | M | H |
[18] | 2009 | Face | The HC based on Bayesian adaption was used to reconstruct face images. | M | H |
[19] | 2010 | Face | According to the global distribution calculated on the user set, the local characteristics of the attacked client are adapted. | M | H |
[20] | 2012 | Face | The HC based on uphill-simplex algorithm was used to reconstruct face images. | M | H |
[21] | 2013 | Face | A simple reconstruction method was proposed based on RBF regression in face eigenspace. | M | H |
[22] | 2014 | Face | Perceptual learning and HC were used to reconstruct real-valued features from the binary template. | M | M |
[23] | 2018 | Face | A Neighbor Deconvolutional Neural Network (NbNet) was proposed to reconstruct face images from deep face templates. | M | M |
[24] | 2010 | Iris | The initial template was divided into blocks of the same size. The pixels in blocks were modified by genetic algorithm. | M | M |
[25] | 2011 | Iris | The texture image was generated from iris template and embedded into a real iris image. | M | M |
[26] | 2013 | Iris | Genetic algorithm was used to reconstruct images from binary templates. | M | M |
[27] | 2020 | Palmprint | Palmprint images were generated by Generative Adversarial Network (GAN) for false acceptance attack. | H | H |
NHD (FNMR = 0) | NHD (EER) | NHD (FMR = 0) | |
---|---|---|---|
PalmCode [36] | 0.425 | 0.370 | 0.330 |
BOCV [37] | 0.450 | 0.390 | 0.365 |
OrdinalCode [38] | 0.440 | 0.340 | 0.285 |
FusionCode [39] | 0.430 | 0.370 | 0.335 |
CompCode [40] | 0.160 | 0.130 | 0.115 |
RLOC [41] | 0.475 | 0.410 | 0.390 |
DOC [42] | 0.465 | 0.420 | 0.400 |
DRCC [43] | 0.445 | 0.390 | 0.360 |
PalmCode [36] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 1.0 | 6.760 | 940 | 23.2 | 3.392 | 0.988 | 0.050 |
Galbally (EER = 0) | 465.6 | 364.91 | 15 | 24.1 | 2.754 | 0.795 | 0.087 |
Galbally (FMR = 0) | 1618.7 | 893.351 | 1 | 22.2 | 1.855 | 0.695 | 0.086 |
MCwN (FNMR = 0) | 1108.3 | 1101.942 | 317 | 35.0 | 3.868 | 0.972 | 0.028 |
MCwN (EER) | 6013.9 | 1791.945 | 4 | 28.6 | 1.635 | 0.866 | 0.045 |
MCwN (FMR = 0) | 9866.2 | 2192.273 | 0 | 26.4 | 1.179 | 0.794 | 0.060 |
BMS (FNMR = 0) | 86.3 | 82.409 | 293 | 48.3 | 5.133 | 0.999 | 0 |
BMS (EER) | 577.1 | 246.770 | 4 | 40.3 | 1.658 | 0.992 | 0 |
BMS (FMR = 0) | 1511.2 | 655.849 | 0 | 36.9 | 1.098 | 0.983 | 0 |
BOCV [37] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 0 | 0 | 0 | 36.1 | 1.574 | 0.973 | 0 |
Galbally (EER = 0) | 1117.8 | 875.181 | 0 | 28.7 | 2.546 | 0.927 | 0.022 |
Galbally (FMR = 0) | 3731.1 | 2743.178 | 0 | 27.9 | 1.894 | 0.913 | 0.020 |
MCwN (FNMR = 0) | 248.5 | 587.795 | 727 | 40.2 | 5.835 | 0.995 | 0.010 |
MCwN (EER) | 8865.1 | 2484.256 | 1 | 29.3 | 1.708 | 0.880 | 0.037 |
MCwN (FMR = 0) | 13,549.7 | 3292.596 | 0 | 27.5 | 1.400 | 0.822 | 0.050 |
BMS (FNMR = 0) | 23.5 | 42.123 | 389 | 54.9 | 7.598 | 0.999 | 0 |
BMS (EER) | 1188.3 | 546.249 | 0 | 34.7 | 1.999 | 0.977 | 0.010 |
BMS (FMR = 0) | 2261.9 | 903.541 | 0 | 32.5 | 1.058 | 0.964 | 0.010 |
OrdinalCode [38] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 0 | 0 | 998 | / | / | 1 | 0 |
Galbally (EER = 0) | 783.5 | 617.210 | 47 | 23.8 | 2.535 | 0.785 | 0.100 |
Galbally (FMR = 0) | 5656.8 | 4221.575 | 0 | 20.6 | 1.392 | 0.579 | 0.094 |
MCwN (FNMR = 0) | 67.7 | 360.238 | 944 | 36.5 | 4.875 | 0.998 | 0.010 |
MCwN (EER) | 9577.8 | 3352.793 | 3 | 26.8 | 1.813 | 0.797 | 0.076 |
MCwN (FMR = 0) | 20,139.2 | 6879.100 | 0 | 23.8 | 1.646 | 0.626 | 0.114 |
BMS (FNMR = 0) | 6.8 | 15.533 | 944 | 41.8 | 4.020 | 1.000 | 0 |
BMS (EER) | 742.8 | 1051.393 | 3 | 30.2 | 2.075 | 0.943 | 0.022 |
BMS (FMR = 0) | 1956.0 | 2222.085 | 0 | 26.9 | 1.135 | 0.895 | 0.026 |
FusionCode [39] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 0 | 0 | 997 | 15.2 | 0 | 1.000 | 0.010 |
Galbally (EER = 0) | 789.0 | 912.764 | 30 | 23.7 | 2.539 | 0.776 | 0.096 |
Galbally (FMR = 0) | 4544.6 | 3667.629 | 0 | 21.2 | 1.554 | 0.617 | 0.093 |
MCwN (FNMR = 0) | 320.6 | 693.742 | 729 | 36.6 | 4.719 | 0.992 | 0.017 |
MCwN (EER) | 6968.7 | 2152.851 | 4 | 28.4 | 1.788 | 0.856 | 0.050 |
MCwN (FMR = 0) | 11660.5 | 2540.917 | 0 | 26.2 | 1.260 | 0.774 | 0.064 |
BMS (FNMR = 0) | 25.4 | 48.533 | 647 | 52.8 | 6.235 | 1.000 | 0 |
BMS (EER) | 952.4 | 692.114 | 3 | 39.5 | 2.185 | 0.991 | 0 |
BMS (FMR = 0) | 2972.4 | 2503.948 | 0 | 36.6 | 1.162 | 0.984 | 0 |
CompCode [40] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Gabally (FNMR = 0) | 0 | 0 | 998 | / | / | 1 | 0 |
Gabally (EER = 0) | 1535.4 | 1323.859 | 7 | 23.1 | 2.717 | 0.736 | 0.114 |
Gabally (FMR = 0) | 7630.8 | 6221.137 | 0 | 20.7 | 1.492 | 0.586 | 0.100 |
MCwN (FNMR = 0) | 5.2 | 41.978 | 976 | 45.8 | 6.306 | 1.000 | 0 |
MCwN (EER) | 6600.8 | 1782.556 | 0 | 27.9 | 1.618 | 0.839 | 0.052 |
MCwN (FMR = 0) | 11,016.4 | 2224.044 | 0 | 25.6 | 1.179 | 0.749 | 0.070 |
BMS (FNMR = 0) | 4.4 | 3.255 | 787 | 54.6 | 3.308 | 1.000 | 0 |
BMS (EER) | 394.6 | 173.592 | 0 | 34.4 | 1.613 | 0.978 | 0.010 |
BMS (FMR = 0) | 1081.9 | 711.593 | 0 | 31.4 | 0.941 | 0.960 | 0.010 |
RLOC [41] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 0 | 0 | 998 | / | / | 1.000 | 0 |
Galbally (EER = 0) | 1094.7 | 845.827 | 9 | 23.6 | 2.587 | 0.763 | 0.098 |
Galbally (FMR = 0) | 2809.6 | 1691.399 | 0 | 22.0 | 1.861 | 0.676 | 0.093 |
MCwN (FNMR = 0) | 16.0 | 99.884 | 959 | 41.4 | 4.549 | 1.000 | 0 |
MCwN (EER) | 4514.2 | 1282.262 | 0 | 29.2 | 1.592 | 0.887 | 0.036 |
MCwN (FMR = 0) | 6416.3 | 1386.419 | 0 | 27.6 | 1.156 | 0.843 | 0.041 |
BMS (FNMR = 0) | 6.2 | 12.336 | 814 | 60.4 | 4.705 | 1.000 | 0 |
BMS (EER) | 1628.5 | 605.921 | 0 | 39.1 | 1.735 | 0.988 | 0 |
BMS (FMR = 0) | 2561.4 | 692.470 | 0 | 37.3 | 1.065 | 0.982 | 0 |
DOC [42] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 0 | 0 | 998 | / | / | 1.000 | 0 |
Galbally (EER = 0) | 973.0 | 831.115 | 12 | 23.7 | 2.628 | 0.767 | 0.097 |
Galbally (FMR = 0) | 3595.0 | 2740.069 | 0 | 21.6 | 1.610 | 0.647 | 0.091 |
MCwN (FNMR = 0) | 159.7 | 396.768 | 741 | 40.9 | 5.285 | 0.996 | 0.010 |
MCwN (EER) | 6569.4 | 1802.938 | 0 | 30 | 1.850 | 0.896 | 0.033 |
MCwN (FMR = 0) | 10,054.7 | 1934.229 | 0 | 28.2 | 1.316 | 0.851 | 0.040 |
BMS (FNMR = 0) | 17.3 | 28.167 | 448 | 56.7 | 6.767 | 1.000 | 0 |
BMS (EER) | 813.6 | 809.699 | 0 | 38.8 | 1.890 | 0.990 | 0 |
BMS (FMR = 0) | 1839.0 | 1896.114 | 0 | 36.4 | 1.078 | 0.984 | 0 |
DRCC [43] | Matching Number (Mean) ↓ | Matching Number (Std) ↓ | PSNR (Inf) ↑ | PSNR (Mean) ↑ | PSNR (Std) ↓ | SSIM (Mean) ↑ | SSIM (Std) ↓ |
---|---|---|---|---|---|---|---|
Galbally (FNMR = 0) | 0 | 0 | 998 | / | / | 1.000 | 0 |
Galbally (EER = 0) | 815.2 | 892.207 | 21 | 23.4 | 2.608 | 0.766 | 0.097 |
Galbally (FMR = 0) | 6313.7 | 4771.276 | 0 | 20.7 | 1.402 | 0.590 | 0.084 |
MCwN (FNMR = 0) | 45.5 | 213.826 | 924 | 41.3 | 6.267 | 0.999 | 0 |
MCwN (EER) | 6841.4 | 1982.581 | 1 | 29.5 | 1.691 | 0.884 | 0.038 |
MCwN (FMR = 0) | 12,168.2 | 2467.356 | 0 | 27.2 | 1.221 | 0.810 | 0.050 |
BMS (FNMR = 0) | 7.3 | 14.359 | 585 | 60.5 | 5.091 | 1.000 | 0 |
BMS (EER) | 693.3 | 373.336 | 1 | 39.3 | 1.905 | 0.991 | 0 |
BMS (FMR = 0) | 2164.1 | 1533.112 | 0 | 36.1 | 1.008 | 0.983 | 0 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 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
Sun, Y.; Leng, L.; Jin, Z.; Kim, B.-G. Reinforced Palmprint Reconstruction Attacks in Biometric Systems. Sensors 2022, 22, 591. https://doi.org/10.3390/s22020591
Sun Y, Leng L, Jin Z, Kim B-G. Reinforced Palmprint Reconstruction Attacks in Biometric Systems. Sensors. 2022; 22(2):591. https://doi.org/10.3390/s22020591
Chicago/Turabian StyleSun, Yue, Lu Leng, Zhe Jin, and Byung-Gyu Kim. 2022. "Reinforced Palmprint Reconstruction Attacks in Biometric Systems" Sensors 22, no. 2: 591. https://doi.org/10.3390/s22020591