Automatic motion tracking of a human in a surveillance video

In this paper we describe a method of automatic motion tracking of a person in a sequence of video frames. The method is suitable for tracking a person walking or running in a surveillance video captured from a single still camera. The method initially removes noise from the captured images then segments the images using frame difference and binary conversion techniques and finally tracks the person using a bounding box based on occurrence of high intensity values horizontally and vertically. Simulation results show that the method can be used for real-time tracking of humans in videos with frame-rate of 25/30 fps. In the context of a Smart-city, the purposed system serves as a part of Internet of Things (IoT) and transmits the capture video to control-center for processing.