Arfina Rahman, Mahesh Banavar
This study introduces a hybrid deep learning model for biometric authentication using low-frame-rate PPG signals, achieving 98% accuracy by combining spatial and temporal features.
Biometric authentication is a way to verify a person's identity using unique biological traits. This study focuses on using photoplethysmography (PPG) signals, which are easy to obtain and measure blood volume changes in the skin, for this purpose. The researchers developed a new method that combines different deep learning techniques to accurately identify individuals from low-quality PPG signals. Their approach was tested on data from 46 people and achieved a high accuracy of 98%, making it a promising solution for secure authentication on mobile devices and wearables.
