AUTOMATED FALL DETECTION SYSTEM FOR ELDERLY PEOPLE BASED ON SENSOR DATA AND VISIONDATA

Falls are one of the most crucial health risks faced by the elderly.Sudden falls may craft serious danger to elderly people and can cause injuries or even longterm disability.By developing an automated fall detection system, the aim is to mitigate these risks and improve the overall well-being of elderly individuals.With the presence of wearable devices, important health parameters can be extracted for healthcare assessment.Along with this, non-wearable devices like cameras also help in monitoring any unexpected fall by capturing and analyzing image data.In this paper, we analyze sensor data and image data as extracted from wearable and non-wearable devices respectively for accurate fall detection.This research emphasizes use of threshold based and machine learning models for sensor data.For image-based fall-detection, fall events are extracted using CNN and YOLO.In summary, the current work offers a promising way for elderly care promoting a safer environment.