结合全局与局部信息的主动轮廓分割模型

针对传统的基于区域的主动轮廓模型在分割灰度不均匀图像和噪声图像存在效果不佳的问题,提出结合全局项与局部项的主动轮廓分割模型。全局项由CV(Chan-Vese)模型的保真项构成,局部项的构建考虑局部区域信息的同时引入反映图像灰度特性的局部熵信息。依据图像灰度的特点,选择合理的全局项和局部项参数,并加入正则项保证曲线在演化过程中保持平滑,保障分割结果的可靠性。通过变分水平集方法最小化能量泛函,依据梯度下降流迭代更新水平集,完成曲线演化。采用模拟图像和实际图像进行实验分析,结果表明,所提出的结合全局项和局部项的主动轮廓模型可以高效地分割噪声严重以及灰度分布不均匀的图像。

An active contour segmentation model combining global and local region-based information

Aiming at the problem of the traditional region-based active contour model, which is difficult to segment the noisy images and intensity inhomogeneous images, an active contour model including global term, local term and regularization term is proposed. Global term is derived from the data fidelity items of Chan-Vese model. Meanwhile, the construction of local term takes into account local intensity information and introduces image local entropy that reflects the grey characteristics. According to the characteristics of different images, reasonable parameters of global item and local item are selected. The regularization term is added to ensure the smoothness of curve evolution and guarantee the reliability of segmentation. A variable level set is used to minimize the energy function to get the gradient descent flow. The experiment results of synthetic images and real images have shown that the approach proposed in this paper is efficient to segment noisy image and inhomogeneity image.