Illumination variation-resistant video-based heart rate monitoring using LAB color space

Abstract The remote photoplethysmography technology based on consumer cameras has been demonstrated to be an effective method for heart rate monitoring. However, artificial signals caused by the ambient illumination variation and facial motion would severely distort the heart rate pulse signal and affect the measurement accuracy of the heart rate. In view of these issues, the conversion from RGB color space to LAB color space is performed to separate the luminance signal, and the smoothness prior approach is employed to remove the stationary artifacts in the raw signals of A channel and B channel. On this basis, the simple combined signal, the difference between A channel and B channel, is introduced to extract the heart rate pulse signal. Finally, the pure signal is decomposed to obtain a set of intrinsic mode functions using the ensemble empirical mode decomposition algorithm, and heart rate is estimated based on the one intrinsic mode function with the highest energy and peak ratio in the range of 0.7 Hz to 3 Hz. To assess the performance of the framework proposed in this paper, experiments in different scenarios are performed and the experimental results show that the method proposed in this paper can effectively estimate heart rate, where the mean absolute bias is 2.59 beats/min (bpm) and the 95% confidence interval is from -7.14 bpm to 3.40 bpm.