Estimation Method for Joint Moment and Muscle Action of Human Leg without Restriction of Work space

康复治疗对防止卧床病人的增加和改善他们的损伤是非常重要的。关于患者的关节力矩或肌肉动作的信息对于医生或PT判断什么样的康复菜单对于每个患者是最佳的是非常有用的。多摄像机系统的三维运动测量和力板的地面反应测量的集成已被广泛用于跟踪人体部位并对其物理行为进行动态分析。然而，这样的方法需要相当大的工作空间，它是非常昂贵的日常康复。在本研究项目中，提出了一种新的估计方法，测量踝关节力矩和反作用力，通过使用种植压力传感器。将压力传感器安装在患者的鞋子中，并通过使用重回归分析来估计用于信号处理的参数（回归系数）。通过使用这种新的估计方法，步态分析和实验结果的垂直反力和脚关节力矩所提出的方法进行比较，以及通过使用传统的方法。研究了压力传感器数量的影响，并且可以看出，该系统可以在超过4个压力传感器的情况下显示相当精确的结果。研究了足部尺寸对测量结果的影响，发现使用同一传感器可以测量多个不同患者的反力和关节力矩，提出了一种简单的三维运动分析方法，该方法只使用一个摄像头，信号处理非常简单。在图像处理中，可以根据标记的面积来估计相机与人体上的标记之间的距离。该系统可以安装在移动设备上。然后，如果我们将该系统安装在步行支持机器上，我们可以在任何地方测量使用步行支持机器进行康复的患者的运动。

Rehabilitation is very important in preventing an increase of bedridden patients and improving their impairments. Information on the joint moment or muscle action of patients is very useful for doctor or PT to judge what kind of rehabilitation menu is optimum for each patient The integration of three dimensional motion measurement with multi camera system and ground reaction measurement with force plate has been widely used to tracking human parts and performing dynamic analysis of their physical behaviors. However such a method needs sizable work space and it is very expensive for daily rehabilitation.In this research project, a new estimation method for measuring ankle joint moment and reaction force by using planter pressure sensors is proposed. Pressure sensors are mounted in the shoes of patient and parameters(regression coefficients) for signal processing is estimated by using heavy regression analysis. By using this new estimation method, gait analysis is carried out and experimental results of the proposed methods on vertical reaction force and foot joint moment compare well with those by using the conventional method. Effect of number of pressure sensors is examined and it can be seen that this system can show considerably precise results with more than 4 pressure sensors. Effect of foot size is examined and it can be seen that reaction force and joint moment of many different patients can be measured by using same sensor mounted shoes.A simple three dimensional motion analysis method by using only one camera and very simple signal processing is proposed. Distance between camera and markers on the human body can be estimated from area of the marker in the image processing. This system can be installed on the moving device. Then if we install this system on a walking support machine, we can measure motion of a patient who uses a walking support machine in his rehabilitation in any place.