Title :
Moving object detection from image sequence in context with multimedia processing
Author :
Nagmode, M.S. ; Joshi, Madhuri A.
Author_Institution :
Pune Univ., Pune
Abstract :
Research in motion analysis has evolved over the years as a challenging field. Moving object detection and tracking is of major interest in applications such as traffic monitoring, military, medicine and biological sciences etc. In this a new algorithm is proposed for the detection and tracking of moving objects. The image sequence is taken from the digital camera. After preprocessing, the first two frames from the image sequence are segmented by using 8 component connectivity. These segments/clusters are used as seeds for the next frame segmentation. From these segments centroid and area of each cluster is calculated. The Kalman estimation is used to estimate the centroid values for the next frame. The moving object is detected and tracked by using Kalman estimated centroid values. Estimated centroid values of the current frame and previous frames are compared and if there is enough change in the centroid position, it means it is the cluster which belongs to the moving object.
Keywords :
image motion analysis; image segmentation; image sequences; multimedia communication; object detection; Kalman estimation; biological sciences; digital camera; image sequence; motion analysis; moving object detection; moving object tracking; multimedia processing; traffic monitoring;
Conference_Titel :
Wireless, Mobile and Multimedia Networks, 2008. IET International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-0-86341-887-7