Activity recognition in smart home using deep learning
Year: 2019
Type: conference-paper
Abstract: Modern societies are facing an important challenge of taking care of the elderly population. With the evolution of the internet of thing, health Smart home is considered as a possible solution to help to maintain their independence and monitoring elder's daily activities. It proposes to provide medicals and caregivers of the real-time status of the person at home and interfere in the emergency situation. One of the main functionalities of health monitoring systems is human activity recognition. In our work, we present a system able to recognize activities of daily living of the person in the smart home based on the deep learning approach. We propose a hybrid neural networks model based on convolutional and on a recurrent approaches using the opportunity dataset for Human Activity Recognition from Wearable Object and Ambient Sensors. We study the activity recognition workflow and the suitability of the hybrid model for different activities of daily living.
Source: HAL
Authors: El Ghazi, Mariam, Aknin, Noura
Institution: Abdelmalek Essaâdi University
Language: French
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4391518234
