Remote PPG Heart-rate Estimation System

In this project, have we developed a real-time demonstration system capable of accurately estimating an individual's heart rate by analyzing changes in light reflection from the person’s skin. This estimation is performed using any standard RGB camera accessible to the user. The estimation is based on the extraction of PPG (Photoplethysmography) signal from the remotely captured video using digital signal and image processing techniques. In the video, we showcase the system’s functionality, demonstrating its operation on a laptop equipped with a built-in webcam. The estimation results are presented alongside the corresponding ground-truth values. For the UBFC-rPPG dataset, our system achieves median of the mean absolute error of 3.8%. Our system’s accurate real-time performance makes it a promising prototype solution for various applications in healthcare, fitness monitoring, and beyond.

An undergraduate project
Performed by: Tamir Malka, Amir Mishael
Supervised by: Hadas Ofir

The Signal and Image Processing Laboratory (SIPL)
Andrew and Erna Viterbi Faculty of Electrical and Computer Engineering
Technion – Israel Institute of Technology

Follow this link to learn more: https://sipl.eelabs.technion.ac.il/