Activities Detection Based On Automatic Area Segmentation For Elderly Solitaries
Chiao-Wen Kao, Shyi-Huey Wu, Bor‐Jiunn Hwang · 2018
The use of wireless signals to detect events such as fall and get up in the home for elderly solitaries is an attractive issue. This paper studied the issue concerned with position and motion of the elderly solitary in the home by using a fixed-wireless transmitter and a receiver on the companion robot where there is no-sensor on the elderly solitary. The main contribution includes proposed the M-DBSCAN (Modified Density-Based Spatial Clustering of Applications with Noise) method to divide home space into several regions automatically and to recognize specific activities based on location-aware. First to estimate the location of elderly solitary in the home, and then to identify specific activities which can improve the accuracy.