作者: Sara Khalifa , Mahbub Hassan , Aruna Seneviratne
DOI: 10.1109/ICMU.2014.6799049
关键词:
摘要: In large shopping malls and airports, pedestrians often change floors using conveniently located lifts escalators. Floor changing activity recognition (FCAR) therefore can be a vital aid to multi-floor pedestrian navigation systems. The focus of this paper is achieve accurate FCAR with the minimal number features. Using experimental data, we compare performance various feature selection methods classifiers trained detect whether user an escalator or lift. results show that accelerometer embedded in smartphone 94% accuracy only 5