| 一种基于多相位水平集的图像分割 |
| |
| 引用本文： | 易欣,贾振红,覃锡忠,杨杰,胡英杰. 一种基于多相位水平集的图像分割[J]. 激光杂志, 2013, 0(6): 37-39 |
| |
| 作者姓名： | 易欣 贾振红 覃锡忠 杨杰 胡英杰 |
| |
| 作者单位： | [1]新疆大学信息科学与工程学院,乌鲁木齐830046 [2]上海交通大学图像处理和模式识别研究所,上海200240 [3]新西兰奥克兰理工大学知识工程与开发研究所新,西兰奥克兰1020 |
| |
| 基金项目： | 教育部促进与美大地区科研合作与高层次人才培养项目. |
| |
| 摘 要： | 基于水平集的方法已经广泛的使用于图像分割中，但是其复杂的计算限制了其应用范围。为此，提出了一种基于水平纂方法的多坦位的c—v模型。该方法首先通过一种新的初始化方法，获得初始轮廓，加快了算法的收敛速度；同时，引用了距离约束项来消除传统c-V模型的重新初始化问题，提高了算法的计算效率；最后，为了获得了更好的边缘定位能力，在模型中加入了梯度信息。分别对模拟合成图像和一般真实图像进行了实验，结果表明该模型具有边界定位准确和收敛速度快等优点。
|
| 关 键 词： | 水平集 c—v模型 多相位 梯度信息 图像分割 |
A multiphase Level set model for image segmentation |
| |
| Affiliation: | YI Xin,JIA Zhen - hong, QIN Xi - zhong, YANG Jie, Raphael Hu( 1. College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China; 2. Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University, Shanghai 200240, China; 3. Knowledge Engineering and Discovery Research Institute, Auckland University of Technology,Auckland 1020,New Zealand) |
| |
| Abstract: | Level set methods have been widely used in image segmentation, due to the heavy computation, its application held has been restricted, to solve it, a new muhiphase level set method which is based on C - V model is proposed. A new initialization method is presented to boost the convergence of the model; a distance regularizing term is adopted to eliminate the need of re - initialization; to better detect the boundaries, gradient information is added into the original model. The model has been applied to synthetic and real images, resuhs show that our model has better ability of edge detection and faster convergence. |
| |
| Keywords: | level set C- V model muhiphase gradient information image segmentation |
| 本文献已被 维普 等数据库收录！ |
|
