Abstract: In this Paper，we propose a new kind of fast image segmentation method with multiple regions based on level set.First，by introducing the new initialization function for the classic level set based image segmentation algorithm，the time cost is dramatically reduced.Then，segmentation of images with multiple regions is completed，with the introduced multiple regions segmentation condition controlling the convergence of the level set function.Our test shows the proposed image multiple regions segmentation method performs well with few time cost.

Key words: level set, image segmentation, curve evolution, energy minimization

摘要： 提出了一种新的基于水平集的图像快速多区域分割方法。首先，在经典水平集分割算法的基础上，通过使用新的水平集初始化函数，有效地改善了水平集分割算法的时间性能；其次，通过引入区域分割控制条件控制水平集函数的收敛过程，实现多区域分割。实验结果表明，提出的多区域分割方法具有较好的分割性能，并且时间耗费少。

关键词: 水平集, 图像分割, 曲线进化, 能量最小化