基于边缘扩散信息拟合的测地线活动轮廓模型

针对测地线活动轮廓模型对轮廓初始化敏感的问题,提出一个基于边缘扩散信息拟合的测地线活动轮廓模型.首先定义了一个与图像边缘法线方向的二阶导数相关的扩散方程,通过求解这个扩散方程获得边缘扩散信息,利用这种边缘扩散信息构建了一个新的力场;然后由该力场驱动活动轮廓演化,使活动轮廓可以从边缘的两侧向真实目标边缘逼近,最终收敛到期望的边缘.本文模型采用一种快速有效的数值方法实现,水平集函数在整个演化过程中无需重新初始化,活动轮廓演化速度得到显著提高.一系列的人工和真实图像的实验结果表明,本文模型不仅对于初始轮廓的位置选择不敏感,并且可以分割弱边界目标、具有复杂几何结构的目标和带有孔洞结构的目标,综合性能优于一些传统算法.

Geodesic Active Contour Model Based on Edge Diffusion Information Fitting

Aiming at the sensibility of geodesic active contour model to contour initialization， a model was proposed based on edge diffusion information fitting.A diffusion functional was defined， which is related to the second derivative in the normal direction of image edge，and the edge diffusion information was obtained by solving this diffusion functional.Then， the edge diffusion information was utilized to construct a novel force field， which drives the active contour to evolve and converge to desirable edges. An efficient numerical method was used for the implementation of the proposed model in order to converge rapidly and avoid reinitialization. Experimental results on a series of real and synthetic images demonstrate that the proposed model is robust to the initial contour， and it can segment the objects with the weak edges successfully， as well as the objects with complex geometry shapes and the objects with interior and exterior boundaries. It has better segmentation performance compared with some traditional models.