一种新的曲线演化混合模型图像分割算法

A new image segmentation algorithm based on hybrid curve evolution model

本文在Mumford-Shah模型的基础上,将传统几何曲线演化的驱动力(图像梯度局部信息)、Mumford-Shah模型的全局信息以及水平集的符号距离函数统一在一个变分框架之下,完成曲线演化过程的数值计算。本混合模型无需重新计算演化曲线的初始位置,可选择较大时间步长。实验结果表明,新的混合模型既保留了原有曲线演化模型的优势,又能高效稳健、快速地完成曲线演化过程。

This paper presents an improved Mumford-Shah model on combining image local gradient and penalizing energy for image segmentation. The Mumford-Shah functional is modified so that the energy depends on the image global information and the image local gradient information. The introduction of the penalizing energy enables us to select larger time step in implementation. Experimental results show that our improved model is superior to the conventional one both in terms of speed and robustness.