LabVIEW环境下基于Kinect的人体姿态识别中手臂动作辨识的探究  

Research on arm motion recognition in human gesture recognition based on kinect in labview environment

作　　者：刘卫玲 刘畅 张耀尹 常晓明 杨玲珍 Liu Weiling;Liu Chang;Zhang Yaoyin;Chang Xiaoming;Yang Lingzhen(College of Data Science,Taiyuan University of Technology,Taiyuan 030024,China;College of Mathematics,Taiyuan University of Technology,Taiyuan 030024,China;College of Information and Computer Science,Taiyuan University of Technology,Taiyuan 030024,China;College of Physics and Optoelectronics,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学大数据学院,太原030024 [2]太原理工大学数学学院,太原030024 [3]太原理工大学信息与计算机学院,太原030024 [4]太原理工大学物理与光电工程学院,太原030024

出　　处：《电子测量技术》2019年第23期153-157,共5页Electronic Measurement Technology

基　　金：国家自然科学基金项目(61575137)

摘　　要：为探究Kinect在人体姿势识别中对手臂动作辨识的分辨率大小,设计并搭建了一种分辨率的测试平台。引入最小二乘法对Kinect骨骼图像进行处理,从中提取出人体手臂的倾斜角度,利用模型人代替真人,作为Kinect的测试对象。采用正态分布的3δ原则,计算出每组测试数据的抖动幅度。根据抖动幅度的分布情况,分析得出Kinect分辨率的大小。经测试,得出了Kinect最佳识别范围,及对人体手臂动作辨识的分辨率,其大小为2°。该数据在体感控制系统的姿态设计中具有一定的参考价值。To explore the resolution of Kinect for arm motion recognition in human gesture recognition,a resolution testing platform was designed and built.Kinect bone image was processed by introducing least square method,and the tilt Angle of human arm was extracted from the Kinect bone image.The model person was used instead of a real person as the test object for Kinect.Using the 3δprinciple of normal distribution,the jitter amplitude for each test case was calculated.According to the distribution of jitter amplitude,the value of Kinect resolution was analyzed and obtained.Through testing,the best recognition range of Kinect and the resolution of human arm movement recognition are obtained,which is 2°in size.This data has certain reference value in the attitude design of the somatosensory control system.

关 键 词：KINECT 分辨率 最小二乘法 3δ原则 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]