[1]张伟伟,薄 华,王晓峰.多特征-谱聚类的SAR图像溢油分割[J].智能系统学报,2010,5(06):551-555.
 ZHANG Wei-wei,BO Hua,WANG Xiao-feng.SAR oil spill image segmentationbased on a multispectral clustering algorithm[J].CAAI Transactions on Intelligent Systems,2010,5(06):551-555.
点击复制

多特征-谱聚类的SAR图像溢油分割(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第5卷
期数:
2010年06期
页码:
551-555
栏目:
出版日期:
2010-12-25

文章信息/Info

Title:
SAR oil spill image segmentationbased on a multispectral clustering algorithm
文章编号:
1673-4785(2010)06-0551-05
作者:
张伟伟薄 华王晓峰
上海海事大学 信息工程学院,上海 200315
Author(s):
ZHANG Wei-wei BO Hua WANG Xiao-feng
School of Information Engineering, Shanghai Maritime University, Shanghai 200135, China
关键词:
合成孔径雷达灰度共生矩阵谱聚类溢油分割
Keywords:
synthetic aperture radar graylevel cooccurrence matrices spectral clustering spill oil segmentation
分类号:
TP751
文献标志码:
A
摘要:
经典的K聚类算法,并不适合实现任意形状的聚类,而且有容易陷入局部最小值的不足.提出基于多个纹理特征的谱聚类算法,该方法用灰度共生矩阵(GLCM)提取合成孔径雷达 (SAR)图像的多个特征值,构建谱聚类的特征矩阵,并依据规范切准则,用K均值聚类的方法对拉普拉斯矩阵的第二小的特征值对应的特征向量进行聚类,实现基于SAR图像的溢油的分割.新方法与传统的K聚类方法比较,可以减少相干斑噪声对分割结果的影响,较好的保持图像边缘.仿真结果显示,该算法对于相干斑噪声影响较大的图像具有较强的鲁棒性.
Abstract:
The classic Kmean clustering algorithm is not suitable for the circumstances of arbitrary shapes and is prone to use the local minimum. In order to fix these shortcomings, a spectral clustering algorithm based on multitexture characteristics was proposed. The algorithm first used graylevel cooccurrence matrices (GLCM) to extract three features of the synthetic aperture radar (SAR) image and construct a characteristic matrix of spectral clustering. Next using the Ncut (Normalizedcut) criterion, it clustered the eigenvector corresponding to the second small eigenvalue of the Laplacian matrix in order to carry out the SAR oil spill image segmentation. Compared with the classic Kmean algorithm, the proposed method reduces the influence of coherent scattering noise on the segmentation result and efficiently conserves the edge of the image. The simulation results also show that the new method has a strong robustness for an image badly affected by the coherent scattering.

参考文献/References:

[1]ARDHUIN F G, MERCIER G, GARELLO R. Oil slick detection by SAR imagery:potential and limitation. Digital Object Identifier 10.1109/OCEANS.2003.178539:164169.
[2]FUMIAKI T, TSUJI S. Computer analysis of visual textures[M]. Boston: Kluwer Academic Publishers, 1990: 310.
[3]HARALICK M. Statistical and structural approaches to texture[C]//Proceedings of the IEEE,1979, 67(5): 786804
[4]郭巍, 张平,朱良,陈曦. 基于AdaBoost.ECOC的合成孔径雷达图像目标识别研究[J]. 哈尔滨工程大学学报,2010, (2): 232236. 
 GUO Wei, ZHANG Ping, ZHU Liang, CHEN Xi. Research on synthetic aperture radar image target recognition based on AdaBoost.ECOC[J]. Journal of Harbin Engineering University, 2010, (2): 232236.
[5]ZHANG Fengli, SHAO Yun, TIAN Wei, et al. Oil spill identification based on textural information of SAR Image[C]//Proc of the IEEE International Geoscience and Remote Sensing Symposium. Boston,USA, 2008: 13081311.
[6]马秀丽,焦李成.基于分水岭谱聚类的SAR图像分割[J].红外与毫米波学报, 2008, 27(6): 452456.
MA Xiuli, JIAO Licheng. SAR image segmentation based on watershed and spectral clustering[J]. Journal of Infrared and Millimeter Waves, 2008, 27(6): 452456.
[7]焦李成,张向荣,侯彪,等.智能SAR图像处理与解译[M].北京: 科学出版社, 2008: 409414.
[8]BACH F R, JORDAN M I. Blind one microphone speech separation: a spectral learning approach[C]//Proc of Nineteenth Annual Conference Neural Information Processing Systems(NIPS). Vancouver, Canada, 2005: 17.
[9] MALIK J, BELONGIE S, LEUNG T, et al. Contout and texture analysis for image segmentation[J]. International Journal of Computer Vision, 2001, 43(1): 727.
[10]FIEDLER M. Algebraic connectivity of graphs[J]. Czechoslovak Mathematical Journal, 1973, 23: 298305.[11]SHI J B, MALIK J. Normalized cuts and image segmentation[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2000, 22(8): 888905.
[12]DONATH W E, HOFFMAN A J. Lower bounds for partitioning of graphs[J]. IBM Res Develop, 1973, 17: 420425.[13]HAGEN L, KAHNG A B. New spectral methods for ratio cut partitoning and clustering[J]. IEEE Trans on Computed Aided Design, 1992, 11: 10741085.
[14]DING C H Q, HE Xiaofeng, SIMON H D, ett al. A minmax cut algorithom for graph partitioning and data clustering[C]//Proc of the 2001 IEEE International Conference on Data Mining. Calofornia, USA, 2001: 107114.
[15]HARALICK R M, SHANMUGAM K, DINSTEIN I. Texture features for image classification[J]. IEEE Trans on Systerms, Man, and Cybernetics, 1973, 3(6): 610621.

备注/Memo

备注/Memo:
收稿日期:2009-11-12.
基金项目:上海市教育委员会科研创新资助项目(08Y2110).
通信作者:张伟伟.E-mail:zhangyuwei1983@126.com.
作者简介:
张伟伟,女, 1983年生,硕士研究生,主要研究方向为图像处理与视频分析. 
 薄 华,女,1971年生,硕士生导师,主要研究方向为遥感图像处理、模式识别、人工智能. 先后主持上海市教委项目1项,参加国家“863”计划和国家自然科学基金项目多项,其中作为第二完成人获军队科技进步奖三等奖2次.发表学术论文10余篇.
王晓峰,男,1958年生, 教授,博士生导师,上海海事大学信息工程学院院长,上海海事大学学报编委,International Journal of Granular Computing, Rough Sets and Intelligent Systems (IJGCRSIS)编委;兼任中国人工智能学会机器学习专业委员会常务委员、智能交通专业委员会委员、粗糙集与粒度计算专业委员会委员;中国仪器仪表学会微机应用学会常务理事;上海海洋与湖沼学会海洋信息技术专业委员会副主任等.任2006、2007 IEEE控制与机器学习国际会议程序委员会委员.担任2008IEEE International Conference on Granular Computing 程序委员会委员,2007 International Conference on Risk Analysis and Crisis Response 会议副主席;国家科技部国际科技合作项目评审专家.主要研究方向为数据挖掘和机器学习.发表学术论文多篇.
更新日期/Last Update: 2011-03-03