The Practice of Business Statistics Full & CD-Rom & Excel Manual & CD-Rom
The Practice of Business Statistics Full & CD-Rom & Excel Manual & CD-Rom
Accurate activity recognition in a home setting
UbiComp '08 Proceedings of the 10th international conference on Ubiquitous computing
SmartBuckle: human activity recognition using a 3-axis accelerometer and a wearable camera
Proceedings of the 2nd International Workshop on Systems and Networking Support for Health Care and Assisted Living Environments
IEEE Transactions on Information Technology in Biomedicine
Understanding challenges and opportunities of preventive blood pressure self-monitoring at home
Proceedings of the 31st European Conference on Cognitive Ergonomics
Hi-index | 0.00 |
Fall injury is a health-threatening incident that may cause instant death. There are many research interests aimed to detect fall incidents as early as possible. Fall detection is envisioned critical on ICT-assisted healthcare future. In this paper, we study fall indicators from smartphone perspective. We use smartphone built in accelerometer and orientation sensor to recognize various falls. We are subsequently to propose a semi-supervised algorithm design for mobile phone as a part of our ongoing project on fall detection application and system with smartphone. The genuine fall event will be evaluated based on multiple indicators as the reference to fall detection. We firstly obtain 1st supervised algorithm with indicators as our features trained using decision tree. After several phases of observation, 1st algorithm becomes our reasoning basis that helps us to propose another accurate, robust, and ideal semi-supervised algorithm for smartphone. The result of our experiment is presented as profiles of various fall characters being used as observation basis in 2nd algorithm. Our experiment shows that our study is promising to be the base of newer algorithm. It is feasible and accurate to detect a genuine fall event with smartphone.