[1]曹琳,云挺,舒华忠.基于曲线波的超声图像分割[J].东南大学学报(自然科学版),2012,42(3):419-423.[doi:10.3969/j.issn.1001-0505.2012.03.005]
 Cao Lin,Yun Ting,Shu Huazhong.Ultrasound image segmentation based on curvelet[J].Journal of Southeast University (Natural Science Edition),2012,42(3):419-423.[doi:10.3969/j.issn.1001-0505.2012.03.005]
点击复制

基于曲线波的超声图像分割()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
42
期数:
2012年第3期
页码:
419-423
栏目:
计算机科学与工程
出版日期:
2012-05-20

文章信息/Info

Title:
Ultrasound image segmentation based on curvelet
作者:
曹琳 云挺 舒华忠
东南大学影像科学与技术实验室, 南京 210096
Author(s):
Cao Lin Yun Ting Shu Huazhong
Laboratory of Image Science and Technology, Southeast University, Nanjing 210096, China
关键词:
Riemann-Liouville分数阶微分 曲线波变换 Adaboost 超声图像 分割
Keywords:
Riemann-Liouville fractional differential curvelet transform Adaboost ultrasound image segmentation
分类号:
TP391
DOI:
10.3969/j.issn.1001-0505.2012.03.005
摘要:
为了提高前列腺超声图像分割的准确率,提出一种基于曲线波的半监督超声图像自动分割方法.首先,采用对微小波动敏感度高的Riemann-Liouville(RL)分数阶微分算子,突出模糊边界并增强超声图像的纹理; 其次,运用曲线波变换对超声图像进行频域中的分解,获得不同子带分量以表达超声图像特征; 然后,基于Adaboost的分类算法识别出超声图像中的病灶区和非病灶区; 最后,采用中值滤波和腐蚀的方法使病灶区域边缘完整、平滑.实验表明,与运用共生矩阵及二进小波作纹理分析的分割结果比较,所提出的方法在准确率上有了很大的改进,分割超声图像效果更佳.
Abstract:
In order to improve the accuracy of prostate ultrasound image segmentation, a semi-supervised automatic segmentation method based on curvelet transform is proposed. First, the Riemann-Liouville(RL)fractional differential operator which is sensitive to the tiny fluctuations is used to enhance the fuzzy boundary and image texture. Secondly, the image is transformed into curvelet domain and different subbands are obtained to represent the ultrasound image characteristics. Thirdly, the Adaboost algorithm is applied to identify the lesion and non-lesion regions in the ultrasound image. Finally, the median filter and the erosion operator are used to smooth the lesion regions’ edge. Experiments show that the proposed method outperforms the approaches based on co-occurrence matrix and dyadic wavelet in terms of accuracy.

参考文献/References:

[1] Yu Y,Molloy J A,Acton S T.Segmentation of the prostate from suprapubic ultrasound images[J].Med Phys,2004,31(12):3474-3484.
[2] Abolmaesumi P,Sirouspour M R.An interacting multiple model probabilistic data association filter for cavity boundary extraction from ultrasound images[J].IEEE Trans Med Imag,2004,23(6):772-784.
[3] Starck J L,Elad M,Donoho D.Redundant multiscale transforms and their application for morphological component separation[J].Adv Imaging Electron Phys,2004,132(1):287-348.
[4] Candès E J,Donoho D L.Ridgelets:a key to higher-dimensional intermittency[J].Phil Trans R Soc Lond A,1999,357(1760):2495-2509.
[5] Binh N T,Thanh N C.Object detection of speckle image base on curvelet transform[J].ARPN Journal of Engineering and Applied Sciences,2007,2(3):14-16.
[6] Murtagh F,Starck J L.Wavelet and curvelet moments for image classification:application to aggregate mixture grading [J].Pattern Recognition Letters,2008,29(10):1557-1564.
[7] Chen B J,Xiao Y,Yu J G,et al.Ultrasonic speckle suppression based on a novel multiscale thresholding technique[C] //International Symposium on Visual Computing.Rabat,Morocco,2010:1-5.
[8] Eltoukhy M M,Faye I,Samir B B,et al.Breast cancer diagnosis based on texture feature extraction using curvelet transforms[C] //International Congress on Instrumentation and Applied Sciences.Kuala Lumpur,Malaysia,2010:1-4.
[9] Pu Y F,Zhou J L,Yuan X.Fractional differential mask:a fractional differential-based approach for multiscale texture enhancement [J].IEEE Transactions on Image Processing,2010,19(2):491-511.
[10] Gan Z F,Yang H Y.Texture enhancement though multiscale mask based on RL fractional differential[C] //2010 International Conference on Information Networking and Automation.Kunming,China,2010:333-337.
[11] Candès E,Demanet L,Donoho D,et al.Fast discrete curvelet transforms[J].Multiscale Modeling and Simulation,2006,5(3):861-899.
[12] Freund Y,Schapire R E.A decision-theoretic generalization of on-line learning and an application to Boosting[J].Journal of Computer and System Sciences,1997,55(1):119-139.

备注/Memo

备注/Memo:
作者简介: 曹琳(1987—),女,硕士生; 舒华忠(联系人),男,博士,教授,博士生导师, shu.list@seu.edu.cn.
基金项目: 国家重点基础研究发展计划(973计划)资助项目(2011CB707904)、国家自然科学基金资助项目(60911130370)、教育部博士点基金资助项目(20110092110023).
引文格式: 曹琳,云挺,舒华忠.基于曲线波的超声图像分割[J].东南大学学报:自然科学版,2012,42(3):419-423. [doi:10.3969/j.issn.1001-0505.2012.03.005]
更新日期/Last Update: 2012-05-20