A new biometric authentication technology 'tooth-grinding authentication' is proposed following fingerprint authentication and face authentication. Authentication is possible just by wearing an earphone.
A lot of research on biometric authentication technology to identify individuals has been conducted, and in recent years, many smartphones are compatible with biometric authentication systems such as fingerprint authentication and face authentication. Meanwhile, a research team
Ear Wearable (Earable) User Authentication via Acoustic Toothprint
https://doi.org/10.48550/arXiv.2204.07199
TeethPass: Dental Occlusion-based User Authentication via In-ear Acoustic Sensing
https://cis.temple.edu/~yu/research/TeethPass-Info22.pdf
Biometric Authentication by Grinding Your Teeth --Unite.AI
https://www.unite.ai/biometric-authentication-by-grinding-your-teeth/
'Shape of dental arch', 'gap between teeth', 'thickness of enamel', 'size of teeth', etc. vary from person to person. Focusing on the shape of teeth that differs from person to person, the research team set out to develop a technology that identifies an individual from the difference between 'the sound of rubbing teeth against each other' and 'the sound of collision between teeth'.
The research team attached a microphone to a commercially available earphone and passed through the inside of the face from the mouth in order to develop a biometric authentication technology that uses 'the sound of rubbing teeth against each other' and 'the sound of collision between teeth'. We have built a system that detects sounds that reach your ears.
According to the research team, when the constructed system performs actions such as 'rubbing all teeth', 'rubbing some teeth', 'colliding all teeth', and 'colliding some teeth'. It is possible to identify an individual by using the sound generated by the tooth. As a result of actually demonstrating 22 subjects wearing prototype earphones, it is possible to recognize an individual with a probability of 93% by making one sound, and succeeding in individual recognition with a probability of 98.8% by making a sound four times. It was confirmed to do.
Commercially available noise canceling earphones use a microphone to detect ambient noise and reproduce the noise-canceling sound to reduce noise. In addition, some noise canceling earphones have a built-in high-performance processing chip such as ' AirPods Pro '. The research team claims that by using earphones with built-in microphones and processing chips like AirPods Pro, biometrics using tooth sounds can be realized without the need for a dedicated machine.
In addition, since the sound of the teeth passes through the jaw and ear canal by the time it reaches the microphone, it is difficult to make an attack that 'creates a tooth model based on the photo and breaks the certification'.
Related Posts:
