|国家预印本平台
首页|基于变分水平集方法的多相图像分割模型

基于变分水平集方法的多相图像分割模型

Multiphase Image Segmentation Model Based on Variational Level Set Method

中文摘要英文摘要

本文针对多相图像分割问题,提出了一种基于变分水平集方法的图像分割的新模型,其能量泛函基于隐式水平集函数和区域特征函数,能够自动处理曲线拓扑结构的变化,并能自然地扩展为基于各种区域竞争策略的多相图像分割模型。对该能量模型采用变分方法求其梯度最速下降方程,通过演化该方程,最终得到模型的最优解。且为使计算更加准确,本文采用离散形式的水平集函数代替传统的符号距离函数。实验结果表明,该模型具有良好的分割性能。

n innovative model based on variational level set method to segment multiphase images was proposed. Its energy functional is based on both the level set function and the regional characteristics function. It can deal with changes of curves’ topology and be expanded to multiphase image segmentation models which are based on other strategies for region competition as well. The gradient flow of the functional was computed using variational method to get the optimized result of the model. In order for more accurate calculation, discrete definition of level set function was used. Experimental results show that the new model has promising effect on image segmentation.

崔丽娜、魏伟波、李华、石洁、潘振宽

计算技术、计算机技术

多相图像分割变分水平集特征函数

Multiphase SegmentationVariational Level Setharacteristics Function

崔丽娜,魏伟波,李华,石洁,潘振宽.基于变分水平集方法的多相图像分割模型[EB/OL].(2009-02-04)[2025-08-18].http://www.paper.edu.cn/releasepaper/content/200902-26.点此复制

评论