Automated, high-accuracy, remote measurement of heart rate
Arrhythmia is a marked symptom of many cardiovascular diseases (CVDs). The accurate and timely detection of heart rate can greatly reduce the harm of arrhythmia to people. However, it is still a challenge to robustly and remotely measure heart rate in daily life due to the changing environmental conditions during measurement, such as the varying light intensity, the movement of people, and the uncertain distance between the sensor to people. In this study, we propose a method to accurately measure human heart rate within a distance of 4.5 meters under different light intensities by simply using a surveillance camera. After a 20-second color video of a persons hand is captured by the camera, a method based on the Fast Fourier Transform (FFT) algorithm is designed to extract the blood volume pulse wave to calculate the heart rate. According to the comparison between the real heart rate and results measured by electrocardiography (ECG), the proposed method achieves an accuracy of 95.8% when the measurement is performed within a distance of 4.5 meters and 90% when within 5.0 meters. Our experiments show that when the illuminance varies between 100-1000 lux (lighting level indoor), we still get the correct results. Our experiments also demonstrate that the proposed method accurately obtain heart rate even when the light intensity is below 32 lux (300-500 lux in a workplace environment). The methods strong adaptability to changing environmental conditions makes it applicable to many scenarios, such as homes of the elderly, classrooms, and other public spaces.