Research on Non-Contact Heart Rate Measurement Method Based on Self-Optimizing Normalized Least Mean Square Algorithm

The non-contact heart rate measurement method based on imaging photoplethysmography is affected by the interference of ambient light, which will seriously affect the identification and extraction of heart rate signals. In order to solve this problem, this paper proposes a denoising method of heart rate signal based on a self-optimizing normalized least mean square algorithm. According to the trend of light changes in the original heart rate signal and the background signal, this method can automatically select the control factor through the least mean square error based on the normalized least mean square adaptive filter, which can maximize the removal of light interference in the image. Compared with the actual heart rate results using finger clip pulse oximeter, the Person correlation coefficient between the average heart rate measured by the two methods under the condition of drastic changes in light intensity is 0. 95, and the mean absolute percentage error is 2. 16%. The results indicate that the method greatly improves the anti-noise ability of the imaging photoplethysmographic against ambient light, which improves the practicability, reliability, and stability of the imaging photoplethysmographic measurement method.