Technological Solutions for Social Isolation Monitoring of the Elderly: A Survey of Selected Projects from Academia and Industry.

Social isolation is likely to be one of the most serious health outcomes for the elderly due to the COVID-19 pandemic, especially for seniors living alone at home. In fact, two approaches have been used to assess social isolation. The first is a self-reported survey designed for research purposes. The second approach is the use of monitoring technology. The objective of this paper is to provide some illustrative publications, works and examples of the current status and future prospects in the field of monitoring systems that focused on two main activities of daily living: meal-taking activity (shopping, cooking, eating and washing dishes) and mobility (inside the home and the act of going out). These two activities combined seem relevant to a potential risk of social isolation in the elderly. Although current research focuses on identifying only ADLs, we propose to use them as a first step to extract daily habits and risk level of social isolation. Moreover, since activity recognition is a recent field, we raise specific problems as well as needed contributions and we propose directions and research opportunities to accelerate advances in this field.

Posture recognition based on fuzzy logic for home monitoring of the elderly.

We propose in this paper a computer vision-based posture recognition method for home monitoring of the elderly. The proposed system performs human detection prior to the posture analysis; posture recognition is performed only on a human silhouette. The human detection approach has been designed to be robust to different environmental stimuli. Thus, posture is analyzed with simple and efficient features that are not designed to manage constraints related to the environment but only designed to describe human silhouettes. The posture recognition method, based on fuzzy logic, identifies four static postures and is robust to variation in the distance between the camera and the person, and to the person's morphology. With an accuracy of 74.29% of satisfactory posture recognition, this approach can detect emergency situations such as a fall within a health smart home.