Title :
Multi sensor system for automatic fall detection
Author :
Chokemongkol Nadee;Kosin Chamnongthai
Author_Institution :
Dept. of Electronic and Telecommunication Engineering, King Mongkut´s University of Technology Thonburi, Bangkok, Thailand
Abstract :
To reduce elderly falling risk in the area where video surveillance system is unavailable due to privacy reason and so on, we need a monitor system which does no recognize what they are doing. This paper presents an elderly-falling detection system using ultrasonic sensors. The ultrasonic technology-based multi sensors are couple of receiver and transmitter together, which are connected to Arduino microcontroller in order to send the elderly person´s fall related signal using WiFi to the processing unit. The sensors are positioned as an array on the roof and wall in the room. The signal is analyzed to recognize human by sensing distance, and detect the action such as standing, sitting, and falling by pattern matching with the standard templates of top and side signal. In experiments, the proposed system can recognize accuracy of falling detection is 93% approximately.
Keywords :
"Acoustics","Sensor arrays","Senior citizens","Monitoring","Ultrasonic variables measurement","Microcontrollers","Sociology"
Conference_Titel :
Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2015 Asia-Pacific
DOI :
10.1109/APSIPA.2015.7415408