Object Tracking Based on Background Subtraction and Kalman Filtering

Object tracking is well-considered as one of the most important tasks in today’s surveillance system. For this to happen, detection and frame tracking needs to be done first. Video frames from the video helps to identify the object as a part of object detection. The two most used algorithm for object detection is background subtraction and frame difference method. This paper proposes the background subtraction method. Again, Kalman filter is a robust and precise algorithm that is used to estimate the precise location of a moving object. Kalman filter is used in this paper to track the object accurately. Finally, the performance evaluation is done from the results of the parameters Percentage Fit Error and Root Mean Square Position Error. A python code is implemented and the simulated results show that the performance of this model is accurate and satisfactory for a real time video.