One of three adults 65 or older falls every year. Injuries sustained among the elderly because of falls are a major problem worldwide. Demographic predictions of population aged 65 and over suggest the need for telemedicine applications in the eldercare domain. In this paper, we propose an intelligent surveillance system that monitors human activities with a particular interest of the problem of fall detection. To make the motion detection and object tracking fully automatic and robust under different illumination conditions, combination of best-fit approximated ellipse around the face and temporal changes of head position, would provide a useful cue for detection of different behaviors. The system identifies the face, collects data suchas speed of movement, and triggers an alarm on the determination of fall event. Reliable recognition rate of experimental results underlines satisfactory performance of our system.