基于康复训练过程的人体步态分析.pdfVIP

Artificial Intelligence and Robotics Research 人工智能与机器人研究, 2016, 5(2), 41-52 Published Online May 2016 in Hans. /journal/airr /10.12677/airr.2016.52005 文章引用: 贾泽皓, 陈伟海, 王建华, 吴星明, 彭强. 基于康复训练过程的人体步态分析[J]. 人工智能与机器人研究, 2016, 5(2): 41-52. /10.12677/airr.2016.52005 Human Gait Analysis Based on the Process of Rehabilitation Exercise Zehao Jia, Weihai Chen, Jianhua Wang, Xingming Wu, Qiang Peng School of Automation Science Electrical Engineering, Beihang University, Beijing Received: May 7th, 2016; accepted: May 27th, 2016; published: May 30th, 2016 Copyright © 2016 by authors and Hans Publishers Inc. This work is licensed under the Creative Commons Attribution International License (CC BY). /licenses/by/4.0/ Abstract As China gradually entered the aging society, the incidence of stroke, traumatic brain injury, spin- al cord injury and other diseases gradually increased. In the future, the medical needs for rehabil- itation will be very urgent, and the appearance of the rehabilitation robot is a relief. In order to analyze the motion gait for patients in the course of rehabilitation, in this article, we use Kinect-II tracking and gather the gait motion of patients in the process of rehabilitation exercise, and then we achieve the global classification of gait by using multidimensional differential dynamic time warping algorithm based on slope constraints. At the same time, in order to realize the local anal- ysis of the gait motion, we provide a gait analysis method which separates the gait information based on semantic feature. Keywords Kinect, Gait Analysis 基于康复训练过程的人体步态分析 贾泽皓，陈伟海，王建华，吴星明，彭 强 北京航空航天大学自动化科学与电气工程学院，北京 收稿日期：2016年5月7日；录用日期：2016年5月27日；发布日期：2016年5月30日 贾泽皓 等 42 摘 要 随着我国逐步迈入老龄化社会，脑卒中、脑外伤、脊椎损伤等疾病在人群中的发病率逐渐升高，未来社 会对于康复医疗的需求将十分迫切，而康复机器人的出现将在很大程度上缓解这一状况。为了在康复过 程中对患者的运动步态进行分析，本文首先利用微软开发的Kinect二代体感传感器跟踪和采集患者在康 复运动过程中的姿态信息，然后采用经过优化的基于Slope Constraints的坡度加权多维微分动态时间规 整算法来实现步态的全局分类。同时为了实现步态的局部分析，本文提出一