Fall detection analysis using the k-nearest neighbor algorithm

Abstract

A fall detection system enhances the quality of life for elderly individuals by allowing them to live independently without constant care. It is more accurate and cost-effective compared to image-based systems. The system includes two key components: detection, which identifies falls by comparing daily activity data with abnormal sensor values, and communication, which alerts emergency contacts. By using heart rate and oxygen sensors, it can determine whether a fall is conscious or unconscious. Wearable devices, particularly wrist devices, provide accurate data, but current models primarily detect falls without offering additional health information. Future improvements may include wireless data transmission for increased efficiency.