Performance Evaluation of Touchless Fingerprint Recognition: A Comparative Study of SVM VS Decision Trees

Since the encounter of COVID-19 pandemic, different aspects of daily life in early 2020 had got considerably impacted. To control the rate of newly introduced viral infections a range of various measures were recommended worldwide such as the use of facial masks, face shield, enhanced hand hygiene practices etc helped to decrease the spread of pathogens in social gatherings. Nevertheless, these specific measures were creating difficulties in ensuring the reliability of biometric recognition methods, such as voice, facial, and hand-based biometrics. To avoid problems associated with contact/touch – based Biometrics, in this work we have designed an algorithm for touchless fingerprint recognition using HOG features and Machine Learning classifiers. Performance of recognition is evaluated for SVM vs Decision Tree algorithms. The integration of "HOG features with SVM" proves to be more effective in Touchless Fingerprint Recognition domain.