水平集图像分割中重新初始化规避的探索

Avoidance of Re-Initialization in Level Set Image Segmentation

水平集方法有效地解决了图像分割中曲线演化过程中的拓扑变化问题,其实质是水平集方法与模型的结合,以水平集方法来求解模型得到的偏微分方程的方法。要想约束水平集函数在迭代过程中保持为符号距离函数,保证水平集函数的稳定收敛,就必须对SDF重新进行初始化。但是每次都对SDF重新进行初始化,大大增加了计算量,浪费了宝贵的时间,从而大幅降低了曲线的演化速度。一直以来,大家在不断地改进算法,缩短每次初始化所需的时间,但收效甚微。SDF重新初始化的规避,使图像分割时曲线演化速度加快,实验结果表明这种方法是非常有效的,并且具有很强的鲁棒性。重新初始化的规避,减少了计算量,使水平集图像分割法能满足更多的生活、工业应用中的实时性要求。

Level set method has effectively solved the problen of the topology change in the procedure of curve evolution of the image segmentation, and the essence of the image segmentation technique using level set method is the combination of level set methods and theoretical models, and to solve partial differential equations model using level set method. In order to keep approximately the level set function as a signed distance function during the iterative procedure and ensure stable the convergence of the level set function, it is recessary to re - initializate signed distance function. But the time- consuming re- initialization procedure is not necessary and the avoidance of re- initialization procedure will speed up the curve evolution and the image segmentation. Already, the results of some experiments show that some models formulations without re - initializate signed distance function is very effective and has a strong rohusmess. The avoidance of re- initialization of signed distance function reduces the time of calculation and makes the level set method more applicable in life and industry. In this paper, exploration done about avoidance of re- initialization in image segmentation using the level set method was reviewed and outlooked.