Skip to main content
Have a personal or library account? Click to login
Presentation Attack Detection through Fingerprint Biometric using Leveraged Hard Voting Method Cover

Presentation Attack Detection through Fingerprint Biometric using Leveraged Hard Voting Method

Open Access
|Jul 2026

Figures & Tables

graphic/j_ijssis-2026-0023_cv_001.jpg
graphic/j_ijssis-2026-0023_cv_002.jpg
graphic/j_ijssis-2026-0023_cv_003.jpg
Figure 1:

Sample fingerprints of Clarkson2015 dataset1.

Figure 2:

Sample fingerprints of Clarkson2015 dataset2.

Figure 3:

Sample fingerprints of Clarkson2015 dataset3.

Figure 4:

Workflow of proposed method.

Figure 5:

Performance of the models on Clarkson LivDet2015 Dataset1.

Figure 6:

ROC curve for Clarkson LivDet2015 Dataset1.

Figure 7:

Precision–Recall curve for Clarkson LivDet2015 Dataset1.

Figure 8:

Performance of the models on Clarkson LivDet2015 fingerprint Dataset2.

Figure 9:

ROC curve for Clarkson LivDet2015 fingerprint Dataset2.

Figure 10:

Precision–Recall curve for Clarkson LivDet2015 fingerprint Dataset2.

Figure 11:

Performance of the models on Clarkson LivDet2015 fingerprint Dataset3.

Figure 12:

ROC curve for Clarkson LivDet2015 fingerprint Dataset3.

Figure 13:

Precision–Recall curve for Clarkson LivDet2015 fingerprint Dataset3.

Figure 14:

Performance comparison of models with the datasets.

Table 1:

Performance on Clarkson LivDet2015 fingerprint Dataset1

Model/parametersF1-ScorePrecisionRecallBPCERAPCERAccuracy
RF0.888790.799030.986000.996000.001000.80000
XGBOOST0.882720.799030.986000.992000.014000.79040
CATBOOST0.888300.800320.998000.996000.002000.79920
HV Method0.889280.800641.000000.996000.000000.81080

[i] RF, random forest.

Table 2:

Model performance on Clarkson LivDet2015 fingerprint Dataset2

Model/parametersF1-ScorePrecisionRecallBPCERAPCERAccuracy
RF0.902980.868760.940000.710000.060000.83167
XGBOOST0.905420.872140.941330.690000.058670.83611
CATBOOST0.906840.864130.954000.750000.046000.83667
HV Method0.909670.867070.956670.733330.043330.84167

[i] RF, random forest.

Table 3:

Model performance on Clarkson LivDet2015 fingerprint dataset3

Model/parametersF1-scorePrecisionRecallBPCERAPCERAccuracy
RF0.942030.911210.975000.380000.025000.90400
XGBOOST0.934170.905250.965000.404000.035000.89120
CATBOOST0.923660.878270.974000.540000.026000.87120
HV Method0.954880.946900.963000.216000.037000.92720

[i] RF, random forest.

Language: English
Submitted on: Aug 22, 2025
Published on: Jul 13, 2026
Published by: International Journal on Smart Sensing and Intelligent Systems
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2026 Dharmendra Kaushik, Alok Kumar Singh Kushwaha, Vinay Kumar, published by International Journal on Smart Sensing and Intelligent Systems
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.