CENTROID BASED MULTIPLE OBJECT TRACKING WITH SHADOW REMOVAL AND OCCLUSION HANDLING

This paper gives complete guidelines for authors submitting papers for the Journal. Object motion detection technology deals with how to extract moving object from videos and eliminate the background and noise as much as possible. Extracted object may be classified in various categories such as humans, vehicles, birds, floating clouds, swaying tree and other moving objects and object tracking is to monitor and objects spatial and temporal changes during a video sequence, including its presence, position, size, shape, etc. In this paper Median filter is used for the background modeling and for integrating new information into current background, Kalman filter is used. The complete and correct object is detected by removing the shadow using the Normalized Cross Covariance method. Object is classified using RMI motion image. Centroid based tracking is essential for modeling the path and occlusion is handled efficiently. The results are taken for various datasets of different scenarios