| Parkinson’s disease is a neurological disease that affects the quality of life of patients and their families,as it is characterized by uncontrolled trembling of the limbs at rest.This study uses computer vision technology to identify and track the key points of the hand,process the spatial trajectory coordinates of the key points,and obtain their frequency and amplitude characteristics.The processed data conduce to doctors assessing the status of patients with Parkinson’s disease.In this paper,the spatial position tracking of 21 key points on the hand based on computer vision technology,the spatial position information of the key points is obtained and processed and analyzed,and the main research contents are as follows:Firstly,the calibration principle of binocular camera is introduced,the imaging principle of the camera,the binocular calibration experiment is carried out,the internal reference matrix,external reference matrix and distortion coefficient of the binocular camera are obtained using the Zhang Zhengyou calibration method,and the calibration results are applied to the calibration experiment of the camera to prepare for the target detection and target tracking experiment of the binocular camera.Secondly,a method with 21 key points on the hand is used for multi-target detection experiments based on deep learning.The principle and performance of different deep learningbased target detection algorithms are introduced and compared,and according to the actual situation of this experiment,the YOLO v5 algorithm,which is equally good in detection accuracy and detection speed,is finally selected to detect 21 key points on the hand using the YOLOv5 algorithm.Based on the detection of 21 key points by YOLOv5,the DeepSort multi-target tracking algorithm is used to track multiple key points and collect the coordinates of key point spatial locations.Finally,the Hilbert-Huang transform is applied to decompose the original data of hand key point tremor coordinates into multiple IMF components and a residual signal by EMD,and then the Hilbert transform is applied to the first three orders of IMF components containing the main feature information of the original signal,and finally the Hilbert spectrum and instantaneous amplitude are obtained.By observing the information characteristics of the Hilbert transformed signal,it provides auxiliary information for health care professionals to diagnose Parkinson’s disease. |
