[1]陈树越,李 颖,刘佳镔,等.基于活动轮廓模型的图像分割改进算法[J].常州大学学报(自然科学版),2019,31(02):82-87.[doi:10.3969/j.issn.2095-0411.2019.02.011]
　CHEN Shuyue,LI Ying,LIU Jiabin,et al.Improved Algorithm of Image Segmentation Based on Active Contour Model[J].Journal of Changzhou University(Natural Science Edition),2019,31(02):82-87.[doi:10.3969/j.issn.2095-0411.2019.02.011]
点击复制

基于活动轮廓模型的图像分割改进算法()
分享到：

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
第31卷
期数:
2019年02期
页码:
82-87
栏目:
信息科学与工程
出版日期:
2019-03-28

文章信息/Info

Title:
Improved Algorithm of Image Segmentation Based on Active Contour Model
文章编号:
2095-0411(2019)02-0082-06
作者:
陈树越; 李 颖; 刘佳镔; 朱 军; 黄 萍
(常州大学 信息科学与工程学院,江苏 常州 213164)
Author(s):
CHEN Shuyue;  LI Ying;  LIU Jiabin;  ZHU Jun;  HUANG Ping
(School of Information Science and Engineering, Changzhou University, Changzhou 213164, China)
关键词:
LGIF模型;  K-means聚类;  图像分割;  活动轮廓;  Micro-CT
Keywords:
LGIF model;  K-means cluster;  image segment;  active contour;  micro-CT
分类号:
TP 391.41
DOI:
10.3969/j.issn.2095-0411.2019.02.011
文献标志码:
A
摘要:
针对CV(Chan-Vese)模型对低对比度和灰度不均匀图像难以分割,以及LGIF(Local and Global Intensity Fitting)模型初始轮廓曲线位置影响分割速度的问题,提出了一种在LGIF活动轮廓模型的能量泛函中添加图像聚类信息的K-LGIF(K-means-Local and Global Intensity Fitting)模型,其使用被提取图像的轮廓作为初始轮廓,不同于已有算法使用规则的图形作为模型的初始轮廓。实验结果表明,所给出的算法不仅能保证图像分割效果,而且能够减少迭代次数、缩短图像分割时间,所给出的算法模型分别比CV,LBF,LGIF模型的运算效率提高了9.22倍、2.46倍和1.42倍。
Abstract:
To solve the problems of CV model for low-contrast and non-uniform gray image segmentation and the location of the initial contour curves which affect the speed of segmentation, K-LGIF(K-means-Local and Global Intensity Fitting)model were proposed. K-LGIF model improves LGIF model through adding the clustering information in the energy function and using the contour of the extracted image as the initial contour, which is unlike the existing method using the regular pattern as the initial contour of the model. The experimental results show that the presented algorithm can not only enhance the segmental quality, but also increase the speed of segmentation to reduce the segmentation time effectively. The average efficiency of image segmentation by means of K-LGIF model is increased by 12.48, 2.78, 0.56 times relative to CV, LBF and LGIF model respectively.

参考文献/References:


[1] SASHIDA S, OKABE Y, LEE H K. Comparison of multi-label graph cuts method and Monte Carlo simulation with block-spin transformation for the piecewise constant Mumford-Shah segmentation model[J]. Computer Vision and Image Understanding, 2014, 119: 15-26.
[2]张芹, 侯德文. 一种结合区域梯度的自适应CV模型[J]. 计算机应用与软件, 2014, 31(3):214-216.
[3]LI C, KAO C Y, GORE J C, et al. Implicit active contours driven by local binary fitting energy[C]// IEEE Conference on Computer Vision and Pattern Recognition. Washington D C: IEEE Computer Society, 2007:1-7.
[4]LI C, XU C. Distance regularized level set evolution and its application to image segmentation[J]. IEEE Trans on Image Processing, 2010, 19(12): 590-622.
[5]LI W. Active contours driven by local and global intensity fitting energy with application to brain MR image segmentation[J]. Computer Medical Imaging and Graphics, 2009, 33: 520-531.
[6]李敏, 梁久祯, 廖翠萃. 基于聚类信息的活动轮廓图像分割模型[J]. 模式识别与人工智能, 2015, 28(7): 665-672.
[7]PETER Z, BOUSSON V, BERGOT C. A constrained region growing approach based on watershed for the segmentation of low contrast structures in bone micro-CT images[J]. Pattern Recognition, 2008, 41: 2358-2368.
[8]欧文杰, 范立南, 孙申申. 基于形状约束的LBF模型的水平集分割[J]. 计算机应用与软件, 2014, 31(11): 204-206.

备注/Memo

备注/Memo:
收稿日期:2018-11-04。
基金项目:江苏省高等学校大学生创新创业训练计划(201610292023Z); 江苏省产学研前瞻性联合研究项目(BY2015027-24)。
作者简介:陈树越(1963—),男,河北定州人,博士,教授。E-mail: chensy@cczu.edu.cn
更新日期/Last Update: 2019-03-30