| 基于改进C-V模型的超声图像分割方法 |
| |
| 引用本文： | 杨海洋,刘奇.基于改进C-V模型的超声图像分割方法[J].中国测试技术,2007,33(6):99-101. |
| |
| 作者姓名： | 杨海洋 刘奇 |
| |
| 作者单位： | 四川大学电气信息学院,四川,成都,610065 |
| |
| 摘 要： | 图像分割是超声医学图像学中的难题之一。改进的Chan-Vese（C-V）法加入了约束符号距离函数的能量项，避免了演化时候的重新初始化。在改进C-V模型的基础上，首先借用分水岭中的思想，找到分割目标的近似轮廓，并以此轮廓生成符号距离函数，然后采用改进的C-V法进行超声图像分割。实验表明，改进的方法有更高的精准度和对多目标分割的能力。
|
| 关 键 词： | 超声图像分割 C-V模型 多目标分割 活动轮廓 水平集 |
| 文章编号： | 1672-4984（2007）06-0099-03 |
| 修稿时间： | 2007-05-122007-07-19 |
Ultrasound image segmentation method based on improved C-V model |
| |
| YANG Hai-yang,LIU Qi.Ultrasound image segmentation method based on improved C-V model[J].China Measurement Technology,2007,33(6):99-101. |
| |
| Authors: | YANG Hai-yang LIU Qi |
| |
| Affiliation: | School of Electrical Information,Sichuan University,Chengdu 610065,China |
| |
| Abstract: | Image segmentation is a problem of chocardiographie images. In the improved Chan-Vese model, the reconciled signed distance function was added, so there is no need to reinitialize during the evolution. Based on the improved C-V model, the approximate contour of targets which gives birth to signed distance function is found firstly, and then the improved C-V method is used to segment the image. Experimental results show that the improved method is more accurate and capable of segmenting multi-targets. |
| |
| Keywords: | Ultrasound image segmentation C-V model Segmentation of multi-targets Active contour Level set |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
