Non-contact Heart Rate Measurement Based on Fusion Technology

Among the various physiological indicators of human health, the heart rate occupies an important position, which can directly reflect the health of the human body. Therefore, it is particularly necessary to monitor human heart rate. At present, heart rate monitoring is mainly measured by a single infrared camera or optical camera, which extracts the heart rate based on subtle changes in temperature or skin color. Low resolution of the infrared image is prone to distortion. Optical image is easy to lose details when the illumination is not enough. Limited by the imaging principles of infrared and optical images, its measurement accuracy needs to be improved. In response to the above problems, we propose a heart rate measurement method based on data fusion. According to the human heart rate data measured by the infrared camera and the optical camera, the accuracy of heart rate monitoring is improved through multi-feature data fusion and analysis. The fusion model is constructed by using the least square method and neural network, and the optimal fusion mode is selected. Experimental results show that the fusion model based on neural network has a higher accuracy compared with other measurement methods.