A Research:Pulse Signal Detection Method Based On Neck Micro Vibration Video

As an important physiological signal,pulse wave contains a wealth of information about cardiovascular,which provides a key basis for the diagnosis and treatment of cardiovascular diseases.Imaging photoplethysmography is a new non-contact pulse wave detection technology.The signal processing technology is used to extract pulse information from the results which are recorded by ordinary imaging equipment from the colour change or slight vibration of the skin surface.In this paper,a novel pulse wave detection method based on the slight changes in signal phases is proposed.Taking the vibrations and motion features of neck skin into consideration,it extracts pulsatile information from the invisible vibration in a subject’s neck.The main contributions of this paper are summarized as follows.(1)A pulse signal detection algorithm based on the micro-vibration video of the neck is proposed.In order to calculate the signal-to-noise ratio in blocks,the small movements of the neck are amplified by video amplification technology.The sub-blocks with high signal-to-noise ratio are selected as significant regions to reduce the interference of environmental noise.On this basis,the method of extracting the phase spectrum and calculating the phase difference using the complex controllable pyramid decomposition is used to extract the pulse wave waveform from the significant region of the neck pulse pulsation.Finally,it is achieved that pulse signal detection and heart rate calculation in the neck micro-vibration video(2)A pulse wave detection software system on the mobile phone is designed and implemented.First of all,the face and carotid artery were selected as the data collection location in this system.Then,through the front and rear cameras of the mobile phone are used for video recording and data collection of the human face and carotid arteries.Finally,the proposed pulse signal detection algorithm is used to extract pulse waves and calculate heart rate.The test results prove that the software has good robustness and accuracy.By designing video pulse wave extraction experiments in different scenarios and comparing with the current mainstream algorithms,this paper draws the following conclusions: The method in this paper has remarkable accuracy and robustness.At the same time,on the basis of ensuring accuracy,some methods including selecting the neck sub-region and optimizing the algorithm are used to reduce the computational burden of the algorithm.As a result,the detection efficiency has been effectively improved,and the practical application range of the detection system has been expanded...