A new fall detection algorithm based on depth information using RGB-D camera

For elderly, fall will cause physical harms such as soft tissue injury, fracture, head trauma, and so on. Elders will restrict their movement due to fear of fall. In this paper, a new real-time fall detection algorithm using depth information from RGB-D camera is proposed. Compared to the latest works, the proposed algorithm can increase computing speed and improve fall event detection accuracy. The experimental results showed that the accuracy of the proposed algorithm is 97.2% and the error rate is 2.8 on average%.