Human Fall Detection from Depth Images using Position and Velocity of Subject

摘要: Fall detection and notification systems play an important role in our daily life, since human fall is a major health concern for many communities today's aging population. There are different approaches used developing elderly people with special needs such as disable. The three basic include some sort of wearable, non-wearable ambient sensor vision based systems. This paper proposes system on the velocity position subject, extracted from Microsoft Kinect Sensor. Initially subject floor plane tracked frame by frame. joints then to measure respect previous location. confirmed using see if all after abnormal velocity. From experimental results obtained, was able achieve average accuracy 93.94% sensitivity 100% specificity 91.3%.