Real time object detection and tracking: Histogram matching and Kalman filter approach

Abstract

In this paper we present an approach to develop a real-time object tracking system using a static camera to grab the video frames and track an object. The work presents the concepts of histogram matching and absolute frame subtraction to implement a robust automated object tracking system. Once the object is detected it is tracked using discrete Kalman filter technique. The histogram matching algorithm proposed here helps to identify when the object enters the viewing range of the camera and the absolute frame subtraction gives better results even with low quality videos. Such a tracking system can be used in surveillance applications and proves to be cost effective.