A simple method for walking posture analysis using accelerometers

Abstract

Tri-axial accelerometer is a useful sensor. Most devices use it to detect the steps and human's activity. Some devices use it for digital spirit level meter. Since the first generation of iPhone was released on June 29, 2007, modern smart phones have built in accelerometers which promise to enable quantifying minute-by-minute what people do (e.g., walk or sit). However, smart phones are not suitable for detecting the walking posture. This paper uses wearable device with the tri-axial accelerometer to detect the patterns of the steps and situation of humpbacked to determine the walking postures which are divided into eight walking postures. Finally, we collect ten healthy subjects to develop our algorithm and test six healthy subjects. The proposed algorithm to determine the accurate rate of walking postures in comparison with the physiotherapist is 92.98%.