Development and Evaluation of a Contactless Heart Rate Measurement Device Based on rPPG

Abstract

Heart rate (HR) is an essential indicator of human health, especially cardiovascular health. Standard HR measurement devices require direct contact with the skin which can cause discomfort to people with sensitive skin. To address this issue, we design and develop a contactless HR measurement device based on remote photoplethysmography (rPPG) using off-the-shelf components. We propose amplitude-selective filtering (ASF) and color-distortion filtering (CDF) as the preprocessing techniques and plane-orthogonal-to-skin (POS) as the core rPPG algorithm. A benchmark assessment on a public dataset shows that the proposed algorithm yields the best performance with a mean absolute error (MAE) of 4.66±4.96 bpm. Experimental results from an in-house dataset of 15 subjects show that the developed device results in an MAE of 3.74±2.56 bpm. It has also been evaluated under different measurement distances, lighting conditions, and camera resolutions. The results indicate good performance of the developed device and its potential as a contactless HR measurement device.