Applying Neural Network Architecture in a Multi-Sensor Monitoring System for the Elderly

Abstract — One of three adults 65 years or older falls every year. As medical science advances, people can live with better health and alone up to a very advanced age. Therefore, to let elderly people live in their own homes leading their normal life and at the same time taking care of them requires new kinds of systems. In this paper, we propose a multi-sensor monitoring system for the fall detection in home environments. The system, which consists of a webcam and heart rate sensor, processes the data extracted from the two different sub-systems by applying neural network n order to classify the fall event in two classes: fall and not fall. Reliable recognition rate of experimental results underlines satisfactory performance of our system. Keywords-Neural Network; fall detection; heart rate; webcam I