YOLOv5-based fall detection method

Abstract

With the rising elderly population in China, detecting whether an elderly person has fallen is one of the problems that people need to pay attention to today, however, most of the current detection methods are affected by problems such as expensive, vulnerable to environment and not easy to implement. In order to solve the above problems, this paper proposes a fall detection method with YOLOv5s as the basic network model, which first enhances the original image, and then improves the loss function and NMS non-maximal suppression. The final results show that applying this improved algorithm model can effectively perform fall detection.