International Journal of Computer Science and Network Solutions (IJCSNS)
The authors here in propose a moving object segmentation method based on motion information classification by x-means clustering and region segmentation. In the proposed method, a current frame is first segmented by a morphological watershed algorithm. Affine motion information obtaining from three frames at the feature points is classified by x-means clustering. Whether a feature point corresponds to an uncovered background region or an occlusion is then estimated based on the temporal correlation of the clustering results, and incorrect classified feature point is excluded.