Heart rate,as an important physiological sign information,contains various information about the physical and emotional state to reflect the real-time health of the human body.Therefore,accurate monitoring of heart rate indicators is of vital importance for the prevention and diagnosis of cardiovascular diseases.The traditional heart rate measurement adopts the contact detection method,which requires the subject’s skin to be in close contact with the measuring instrument.This requirement imposes certain restrictions on the actual application field.At the same time,prolonged skin contact can easily cause discomfort to the subject,so it is difficult to meet the daily guardianship needs.The non-contact heart rate detection method based on facial video has become a hot topic of research in recent years.The most widely used method of measuring heart rate is Imaging Photoplethysmography(IPPG).This method has the advantages of non-contact,convenience,and low cost,but the measurement results are easily interfered by factors such as the surrounding environment and human motion.At the same time,the current non-contact heart rate detection technology based on facial video focuses on theoretical research and simulation experiments,and there are relatively few systems to meet the actual needs.In response to the above problems,this thesis completes the following research work:(1)On the basis of reading many related research documents,this topic deeply analyzes the principle of imaging photoplethysmography,and studies the improvement and shortcomings of the current mainstream heart rate detection algorithm based on facial video.(2)Aiming at the problem of noise interference in the heart rate detection process,this thesis designs a heart rate detection method based on HSL skin detection and chromaticity model.This method first detects facial skin in the HSL color space to extract clean signal data;then uses the relationship between different color channel data in the chromaticity model to reduce motion interference;finally,the signal is processed by Fourier transform Spectrum analysis for heart rate estimation.At the same time,this topic conducts comparative experiments to verify the accuracy of the method.(3)In view of the current lack of a system designed and developed to meet the needs of medical institutions,this topic relies on the theory of object-oriented technology to design and implement a non-contact heart rate detection system based on facial video.The system combines the heart rate detection method designed in this subject to meet the monitoring and management needs of medical institutions in real-life scenarios.The development process of the system is based on object-oriented technology and software engineering development specifications.At the same time,system testing shows that the overall function has good accuracy and stability. |