BlowPrint: Blow-Based Multi-Factor Biometrics for Smartphone User Authentication

TL;DR

BlowPrint introduces a novel behavioral biometric method using phone blowing acoustics, combined with facial recognition, achieving high accuracy and robustness for smartphone user authentication.

Contribution

This work presents BlowPrint, a new biometric authentication technique based on acoustic patterns from blowing on a phone, enhancing multi-factor biometric security.

Findings

Achieved 99.35% accuracy with blow acoustics

Attained 99.96% accuracy with facial recognition

Combined approach reached 99.82% accuracy

Abstract

Biometric authentication is a widely used security mechanism that leverages unique physiological or behavioral characteristics to authenticate users. In multi-factor biometrics (MFB), multiple biometric modalities, e.g., physiological and behavioral, are integrated to mitigate the limitations inherent in single-factor biometrics. The main challenge in MFB lies in identifying novel behavioral techniques capable of meeting critical criteria, including high accuracy, high usability, non-invasiveness, resilience against spoofing attacks, and low use of computational resources. Despite ongoing advancements, current behavioral biometric techniques often fall short of fulfilling one or more of these requirements. In this work, we propose BlowPrint, a novel behavioral biometric technique that allows us to authenticate users based on their phone blowing behaviors. In brief, we assume that the way users blow on a phone screen can produce distinctive acoustic patterns, which can serve as a unique biometric identifier for effective user authentication. It can also be seamlessly integrated with physiological techniques, such as facial recognition, to enhance its robustness and security. To assess BlowPrint's effectiveness, we conduct an empirical study involving 50 participants from whom we collect blow-acoustic and facial feature data. Subsequently, we compute the similarity scores of the two modalities using various similarity algorithms and combine them through score-level fusion. Finally, we compute the accuracy using a machine learning-based classifier. As a result, the proposed method demonstrates an accuracy of 99.35% for blow acoustics, 99.96% for facial recognition, and 99.82% for the combined approach. The experimental results demonstrate BlowPrint's high effectiveness in terms of authentication accuracy, spoofing attack resilience, usability, non-invasiveness, and other aspects.