An effective segmentation of moving objects by a novel local regions-based level set

Abstract

This paper presents new local regions based level set model for segmenting multiple moving objects in video sequences captured by a stationary camera. The main idea evolves around the reformulation of well-known global energy in local way, by utilizing little squared windows centered on each point over a thin band surrounding the zero level set, hence the object contour can be reshaped into small local interior and exterior regions that lead to compute a family of adaptive local energies. Moreover, we propose to adapt the smoothness of the contours with an automatic stopping criterion. The proposed method has been tested on different real videos, and the experiment results demonstrate that our algorithm can segment effectively and accurately the moving objects.