Elderly people activity monitoring with involved binary sensors and Deep Convolution Neural Network

Data reveal that by 2025, the number of elderly population will have surpassed 1.4 billion. The percent of the elderly desire a healthy and private environment. They might be homebound without even any external help or nursing due to the heavy pricing of elderly treatment centres. The activities of the elderly must be continuously monitored in order to detect indications and symptoms of any sickness. Despite the fact that some elderly people live in shelters, their behaviour is heavily controlled by handheld sensors and webcams. In addition, practically all surveillance activities intrude on the privacy of the elderly. As a result, a Deep Convolutional Neural Network (DCNN) model has been proposed as a means of ensuring perfect anonymity while also monitoring activities. This model was built using the Aruba data set. It was created by collecting inputs from numerous binary sensors throughout the test house and predicting the most likely sequence of activities that the monitored individual would carry out in the least period of time possible. The obtained image data are then processed into information and sent into the Deep Convolution Neural Network model that has been recommended as an input. Twelve basic everyday activities are taken into account when analysing the DCNN. Finally, with an F1 score of 0.82, the proposed Deep Convolution Neural Network model outperforms the present model in all twelve activities, as well as all other key assessment parameters.