基于视频图像边缘检测的人体下肢运动步态

背景：步态规律主要应用人体行走的运动学、动力学等参数进行描述。目前在运动医学、康复工程和仿生学等领域，步态分析可为确定疾病诊断、康复和治疗方案提供重要依据。目的：应用基于人体运动图像的测量装置系统，采集人体在跑步机上行走的下肢运动步态视频，分析人体步态运动规律。方法：采用基于双摄影机的人体运动图像捕捉系统，在人体的左右髋关节、膝关节、踝关节及脚板设置标识点，对人体在跑步机上行走的下肢运动步态视频进行采集。应用图像边缘检测的原理，对测量数据进行了图像处理和分析，得到人体正常步速行走时，左右大腿与竖直方向夹角、小腿与竖直方向夹角、脚板与竖直方向夹角及膝关节、踝关节标识点的关节角度变化规律。结果与结论：基于视频图像边缘检测人体下肢的运动步态，成本相对低廉，数据误差较小，精度与进口设备较接近。应用该测量结果初步构建了人体步态行走数据库，为建立步态评定标准、异常步态判别以及进一步的康复治疗提供了依据。

Normal human gaits based on image edge detection

BACKGROUND:The parameters of kinematics and dynamics are usually used to describe the law of gaits. Up to now, in the fields of sports medicine, rehabilitation engineering and bionics, gait analysis can provide important basis for ascertaining the scenario of diagnosis disease, the scenario of treatment and recovery.OBJECTIVE:To acquire the lower extremity motion gaits video on the treadmill by the image measuring device system based on the human motion in order to analyze the regular pattern of human gaits.METHODS:The identification points were set on the right and left hip joints, knee joints and foot plate by the image-gathering system in order to acquire the lower extremity motion gait while the subject walked on a treadmill. After processing and analyzing the data based on the theory of image edge detection, angle between the right and left thighs and vertical direction, angle between leg and vertical direction, angle between foot plate and vertical direction and the changes of the angles of identification points of knee joints and ankle joints were obtained.RESULTS AND CONCLUSION:The human lower extremity motion gait obtained according to the theory of image edge detection has relatively low cost and small error, and the accuracy was similar to the imported equipment. Finally, a database of human gait was built based on the measurement results, and it provided a basis for the further establishment of assessing standard for human gait, better analysis of abnormal gait and rehabilitation treatments．