基于MIMU的人体三维姿态检测系统设计  

Design of 3D Human Posture Detection System Based on MIMU

作　　者：杨洪兴[1] 王涛[2] 李冰[2] 赵乾[2] 陈鲁[2] 

机构地区：[1]山东建筑大学信息与电气工程学院,山东济南250101 [2]山东建筑大学机器人研究所,山东济南250101

出　　处：《信息技术与信息化》2012年第6期80-83,共4页Information Technology and Informatization

摘　　要：人体三维姿态检测技术在健康监护、人机交互、动作捕捉等领域具有重要的应用价值。传统的检测方法如基于计算机视觉的的动作捕捉识别,识别算法复杂,计算量大,设备成本高,无法广泛推广。本文研究一种基于微惯性测量组合(MIMU)的人体三维姿态检测系统,采用三轴加速度计、三轴陀螺仪、三轴磁传感器,实时检测人体的三维姿态参数,并通过四元数法计算人体姿态坐标。使用OpenGL绘制人体模型,可实时模拟人体三维姿态。The 3D human posture detection is an important technology in the field of health surveillance, human-computer interaction and motion capture, etc. Traditional detection methods such as motion capture and identification based on computer vision have some defects of the complicated recognition algorithm, huge calculation time and expensive equipments. This paper presents a 3D human posture detection system based on MIMU, which consists of three-axis accelerometer, three-axis gyroscope, three-axis magnetic sensor, and the parameters of human posture are detected by MIMU in real-time. Body posture coordinates are calculated in quatemion. Human model is drawn by OpenGL and it can simulate the 3D human posture in real-time.

关 键 词：三维姿态 微惯性测量组合 四元数 OPENGL 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]