基于边缘检测函数尺度变换的水平集图像分割

针对Chan-Vese模型（C-V模型）存在收敛缓慢等缺陷,给出一种基于边缘检测函数尺度变换的水平集图像分割算法。引入边缘检测函数对C-V模型进行改进,在不降低分割质量的前提下,提高图像分割的速度。为了增强改进模型的灵活性,提出对边缘检测函数进行尺度变换的方法。实验结果表明,改进模型有良好的分割效果,尺度变换能有效加快改进模型的演化速度,保持分割过程的稳定性。

Level Set Image Segmentation Based on Scale Transform of Edge Detection Function

To deal with the disadvantages of Chan-Vese model(C-V model), such as convergence slowness, a level set image segmentation algorithm based on scale transform of edge detection function is proposed. This paper improves the C-V model by introducing the edge detection function, and increases the speed of image segmentation without lowering the quality of segmentation. In order to boost up the flexibility of the improved model, a technique of scale transform to act on the edge detection function is proposed. Experimental results show that the improved model can achieve good effect, and the scale transform can effectively speed up the evolution of the improved model and maintain the stability of segmentation process.