Video Surveillance for Child Safety: Fall Detection and Alert System Using Computer Vision

Preventing serious falls is a major problem for a working mother and a caregiver as and when they are keeping a watchful eye to look after their young kids all the time at home. Generally, young children like to climb on everything, including consumer commodities, like tables, highchairs, and cupboards, etc., that are not designed to be climbed on. As a result, children are at risk of severe injuries. Some falls can lead to death or permanent disability. Because of that, Child accidents have become a significant concern in the contemporary world due to the high rate of severe injuries and deaths caused. And the same can be proven by both research results and evidence such as CCTV footage collected from various locations. The implemented research could provide a solution to the explained problem. It helps to both detect and prevent incidents that could negatively affect child safety, such as child falling action on video surveillance footage. We present an approach use real-time video feed and deep learning and computer vision pipeline that detects child fall action to identify potential domestic fall-related child accidents in surveillance footage