A novel wearable smart button system for fall detection
Wei Zhuang, Xiang Sun, Yueyan Zhi, Han Yue, Hande Mao · AIP conference proceedings · 2017
Fall has been the second most cause of accidental injury to death in the world. It has been a serious threat to the physical and mental health of the elders. Therefore, developing wearable node system with fall detecting ability has become increasingly pressing at present. A novel smart button for long-term fall detection is proposed in this paper, which is able to accurately monitor the falling behavior, and sending warning message online as well. The smart button is based on the tri-axis acceleration sensor which is used to collect the body motion signals. By using the statistical metrics of acceleration characteristics, a new SVM classification algorithm with high positive accuracy and stability is proposed so as to classify the falls and activities of daily living, and the results can be real-time displayed on Android based mobile phone. The experiments show that our wearable node system can continuously monitor the falling behavior with positive rate 94.8%.