Real-time Detecting Human Fall Based on Kalman Filter
In the medical field, accelerometers are often used for measuring inclination of body segments and activity of daily living because they are small and require little power. In this paper we propose the detecting human fall based on Kalman filter (DHFK) algorithm. The algorithm implemented by a portable device with MEMS accelerometer. Changes of accelerometer output are continually processed by a simple Kalman filter, to identify still postures, postural transitions, and dynamic movements of human body in real time. According to the filtered result and change in angle of the rotating accelerometer coordinate, the algorithm determinates whether or not there is a fall during the state transition. The method does not require recording the initial coordinate system of accelerometer, so it is easy to wear for users. Experimental results show that DHFK achieves high accuracy in human fall with reasonable cost
Kalman filter fall detection accelerometer
Na Li Yibin Hou Zhangqin Huang Guangli Han
Embedded software and systems institute,Department of embedded technology Beijing University of technology Beijing China
国际会议
太原
英文
260-264
2011-02-26（万方平台首次上网日期，不代表论文的发表时间）