Research Results


應用雙軸加速度計之人體 姿態辨識系統
Remote Posture Monitoring Using a Two-Axis Accelerometer

研 究 生: 王耀慶 Yao-Qing Wang
指導教授:宋開泰 博士
Dr.Kai-Tai So ng

本研究應用無線傳送之加速度訊號發展一套人體姿態辨識系統。此系統包括一雙軸加速度訊號量測、傳輸模組及姿態辨識軟體。其中加速度量測模組由一雙軸加速度計、RF無線傳輸模組及一8-bit微控器所組成。在接收端有RF無線接收模組,通過8-bit微控器與個人電腦或機器人主電腦溝通。在文中文發展小波理論演算法,將訊號經過5層小波分解,得到低頻小波係數。再利用萃取出來的特徵,經過比對辨識,完成上樓、下樓、走路、站和坐五種姿態辨識系統。

In this study, a remote posture monitoring system has been developed for human status estimation. We propose a wearable sensor including a two-axis accelerometer, 8 bit microcontroller and RF wireless communication module. The acquired body motion signals are transmitted to a host computer via RF link for further processing. This thesis develops an algorithm for feature extraction using wavelet transform. We decompose the signal into five levels and extract low-frequency components for useful features. Pattern recognition techniques are applied then to recognize five basic postures: up stairs, down stairs, walking, standing, and sitting. On-line experimental results are presented to show the effectiveness of the proposed method.

圖 1 遠端姿態估測系統架構圖

圖 2 人體姿態訊號擷取傳送模組

Best view with 800x600| maintained by Jotarun Lin | Update Date: