Fine-Grained and Context-Aware Behavioral Biometrics for Pattern Lock on Smartphones
summary
A system uses behavioral data during pattern entry plus a context module that detects body posture and sensor context to authenticate rather than just the drawn pattern It collects 14,850 samples from 77 participants and classifies pattern segments into 12 polyline types then groups them into strong medium and weak segments and applies a polyline weighted overlap scoring scheme Two learning pipelines are evaluated feature engineering and CNN-based representation learning and an OCSVM verifier avoids needing attacker data and yields lower EER than prior methods
AI-written summary · about
chapters (6)
related talks
06mListen to Your Fingers: User Authentication based on Geometry Biometrics of Touch Gesture
06mDBTB INT Vinay Prabhu r
19mStay Cool! Understanding Thermal Attacks on Mobile-based User Authentication
06mTowards Usable and Secure Location-based Smartphone Authentication
23mExploring Intentional Behaviour Modifications for Password Typing on Mobile Touchscreen
45mOverview of Biometric Auth
25mContinuous Biometrics
21mMaking authentication your competitive advantage under GDPR and PSD2