Design Of Joint Range Of Motion Measurement And Evaluation System Based On Kinect

With the continuous development of China's economy and society and people's yearning for a better life,people are paying more and more attention to their own health.The equipment and software platform for human behavior analysis and posture detection have entered the market from the research stage.Range of motion is a common test item in clinical rehabilitation medicine,which can be used as an important evaluation index in postoperative recovery,disability identification and other fields.At present,manual measurement tool,namely protractor,is widely used in clinical practice to measure the range of motion of human joints.Although this method is simple,it is inefficient and wastes a lot of human resources.Moreover,the measurement results are subjective to a certain extent.In order to solve the disadvantages of the traditional manual measurement method,this paper proposes a new measurement method,which uses the latest generation of Microsoft Azure Kinect depth camera to measure the range of motion of human joints,and designs a measurement and evaluation system.This system is designed to replace the measurement of range of motion used in traditional medicine and assist rehabilitation doctors to evaluate and treat patients' rehabilitation.The work of this paper is as follows:The open source skeletal tracking SDK of Azure Kinect can obtain the position and coordinate information of human joints,but there will be data jutter.In this paper,Holt exponential smoothing method is proposed to filter and de-noise the data stream.Experiments have verified that this method has a good smoothing effect on the bone joints data obtained from Azure Kinect.A method for calculating range of motion was designed according to 32 human bone joints obtained by Azure Kinect bone tracking technology.The data of node position coordinates obtained from Azure Kinect open source bone tracking SDK is prone to data jittering.In this paper,Holt exponential smoothing method is proposed to filter and de-noise the node data.Experiments show that this method has a good smoothing effect on the bone node data jittering.Through the research of rehabilitation evaluation methods and clinical investigation,the rehabilitation evaluation index of range of motion is obtained,which can assist rehabilitation doctors in rehabilitation evaluation.By using Azure Kinect technology development platform,the calculation method of range of motion and data jitter smoothing algorithm proposed in this paper,a software system for measuring range of motion of human body is designed in this paper.The system platform is constructed by adopting Electron framework,and various visual plug-ins are used to present data.The Echarts plugin can display the joint Angle value of each frame in the video in the form of broken line graph on the front interface,and its maximum value is the range of motion.In order to check the test accuracy and the best measurement position of the system,this paper designed a comparative test,through a large number of test data to carry out a comparative analysis,and the system of each part of the module function test.The test results show that the application system can replace the traditional method of measuring range of motion.