Noncontact heart rate measurement using a high-sensitivity camera in a low-light environment

Abstract

We propose a method for the remote estimation of the heart rate and heart rate variability spectrogram by analyzing the hemoglobin concentration obtained from RGB facial videos taken in a low-light environment. The monitoring of emotion has potential in areas such as market research, safety, and health. In particular, methods of analyzing the heart rate obtained from RGB video are expected to be used practically. However, these studies cannot be applied in dark locations where monitoring is necessary, such as an infant’s bedroom, a crime-prone road, and within a car. The proposed method, therefore, uses a highly sensitivity RGB camera capable of capturing videos at low illuminance. As the result, we could measure the heart rate with accuracy exceeding 99% and estimate the heart rate variability spectrogram with high accuracy for low-light environments of 10 lx, which corresponds to brightness levels of the monitoring environments given above.