[1]宋省伟,宗静静,邱天爽△,等.基于全局和区域可伸缩拟合局部熵活动轮廓模型的超声图像分割*[J].生物医学工程研究,2019,01:37-42.
SONG Shengwei,ZONG Jingjing,QIU Tianshuang,et al.Ultrasound image segmentation based on global and local entropy (LE)based region-scalable fitting active contour model[J].Journal of Biomedical Engineering Research,2019,01:37-42.
点击复制
基于全局和区域可伸缩拟合局部熵活动轮廓模型的超声图像分割*(PDF)
《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]
- 期数:
-
2019年01期
- 页码:
-
37-42
- 栏目:
-
- 出版日期:
-
2019-03-25
文章信息/Info
- Title:
-
Ultrasound image segmentation based on global and local entropy (LE)based region-scalable fitting active contour model
- 文章编号:
-
1672-6278 (2019)01-0037-06
- 作者:
-
宋省伟1; 宗静静1; 2; 邱天爽1△; 张晓博3
-
1.大连理工大学电子信息与电气工程学部,辽宁 大连 116024; 2.大连交通大学电气信息工程学院,辽宁 大连 116028;3.大连大学附属中山医院,辽宁 大连 116001
- Author(s):
-
SONG Shengwei1; ZONG Jingjing1; 2; QIU Tianshuang1; ZHANG Xiaobo3
-
1.Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian 116024, China;2.School of Electrical & Information Engineering, Dalian Jiaotong University, Dalian 116028;3.Affiliated Zhongshan Hospital of Dalian University, Dalian 116001
-
- 关键词:
-
超声图像分割; 活动轮廓模型; 区域可伸缩拟合局部熵; 水平集方法; Chan-Vese模型
- Keywords:
-
Ultrasound image segmentation; Active contour model; Region-scalable fitting based on local entropy; Level set method; Chan-Vese model
- 分类号:
-
R318
- DOI:
-
10.19529/j.cnki.1672-6278.2019.01.08
- 文献标识码:
-
A
- 摘要:
-
针对区域可伸缩拟合局部熵(region-scalable fitting based on local entropy,RSF_LE)模型图像分割效率低的问题,本研究提出一种改进的RSF_LE模型。定义带有加权局部灰度拟合项以及辅助的加权全局灰度拟合项的能量泛函,其中加权局部灰度拟合项负责对目标边界附近的轮廓进行诱导,使其靠近目标物边界,加权全局灰度拟合项利用图像的全局信息来引导远离目标的轮廓向目标靠拢,该方法可以克服传统的RSF_LE模型分割算法效率低下的问题,并提高了该方法的鲁棒性。
- Abstract:
-
For the low computation efficiency of the region-scalable fitting-local entropy(RSE_LE) CV model image,a new algorithm called WLGIF which combines weighted local intensity fitting and global intensity fitting was proposed in this paper. Weighted local intensity fitting term was responsible for inducing a local force to attract the contour and made it stop when reaching the object boundaries. Weighted global intensity fitting term pushed the contour far away from object boundaries into the object. Combination of global and local intensity fitting terms promoted the efficiency of computing since global information was considered. Ultrasonic medical images were used for image segmentation experiments. Compared with other algorithms, our new algorithm has a better promotion of computation efficiency and is also robust to intensity inhomogeneity.
备注/Memo
- 备注/Memo:
-
(收稿日期:2018-10-07)国家自然科学基金资助项目(61671105);辽宁省自然科学基金指导计划项目(20170540127);辽宁省教育厅科学研究项目(JDL2016024)。△通信作者Email: qiutsh@dlut.edu.cn
更新日期/Last Update:
2019-05-24