A robust adaptive method for detection and tracking of moving objects

Abstract

The major difficulty in any object tracking system is to detect the moving objects efficiently in varying environment. This paper presents a robust moving object detection method in videos and discusses its applications to human and vehicle detection. Our method consists of average background model with supportive secondary model and an adaptive threshold selection model based on Gaussian distribution. The average background model is used for background modelling as used in [Narayana, 2007] and the background subtraction system is used to provide foreground image through difference image between current image and model image. The adaptive threshold method is used to simultaneously update the system to environment changes. This method is tested on various environments and experimental results show that proposed method is more robust and efficient than others in video-based object detection and tracking.