[1]王立国,马赫男,赵亮,等.基于改进K_Medoids算法的高光谱图像聚类[J].哈尔滨工程大学学报,2018,39(09):1574-1581.[doi:10.11990/jheu.201706108]
 WANG Liguo,MA Henan,ZHAO Liang,et al.Hyperspectral image clustering based on improved K_Medoids algorithm[J].hebgcdxxb,2018,39(09):1574-1581.[doi:10.11990/jheu.201706108]
点击复制

基于改进K_Medoids算法的高光谱图像聚类(/HTML)
分享到:

《哈尔滨工程大学学报》[ISSN:1006-6977/CN:61-1281/TN]

卷:
39
期数:
2018年09期
页码:
1574-1581
栏目:
出版日期:
2018-09-05

文章信息/Info

Title:
Hyperspectral image clustering based on improved K_Medoids algorithm
作者:
王立国 马赫男 赵亮 石瑶
哈尔滨工程大学 信息与通信工程学院, 黑龙江 哈尔滨 150001
Author(s):
WANG Liguo MA Henan ZHAO Liang SHI Yao
College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China
关键词:
高光谱K_Medoids算法Canopy算法等距映射算法测地距离聚类
分类号:
TP753
DOI:
10.11990/jheu.201706108
文献标志码:
A
摘要:
为了解决在复杂的、数据量庞大的高光谱图像中汇集出参考价值较高的聚类组合问题,本文提出一种基于流形的K_Medoids改进算法并应用于高光谱图像的聚类实践中。该算法应用改进的Canopy算法进行初值选定,通过基于流形的测地距离所生成的像元距离矩阵来完成K_Medoids算法的聚类过程。该算法对传统聚类算法所具有的一些难以解决的弊端起到良好的抑制作用。利用AVIRIS图像对该算法进行验证,实验结果表明:与传统方法相比,该算法在类内距离、类间距离、Jaccard系数、Rand系数,以及聚类图像的直观对比五个评价标准下能够取得比传统方法更好的效果。

参考文献/References:

[1] 李翔. 高光谱影像的聚类分析及应用[D]. 北京:北京交通大学, 2015. LI Xiang. Clustering analysis of hyperspectral image and its application[D]. Beijing:Beijing Jiaotong University, 2015.
[2] 李波. 高维数据的流形学习分析方法[M]. 武汉:武汉大学出版社, 2016. LI Bo. Manifold learning analysis method for high dimensional data[M]. Wuhan:Wuhan University, 2016.
[3] YUE Xia, WANG Man, YUE Jun, et al. Parallel K-Medoids++ spatial clustering algorithm based on MapReduce[J]. arXiv preprint arXiv:1608.06861, 2016.
[4] AGHAEE A, GHADIRI M, BAGHSHAH M S. Active distance-based clustering using K-Medoids[M]//BAILEY J, KHAN L, WASHIO T, et al. Advances in Knowledge Discovery and Data Mining. Cham:Springer International Publishing, 2016.
[5] 曹丹阳, 杨炳儒, 李广原, 等. 一种基于CF树的K-Medoids聚类算法[J]. 计算机应用研究, 2011, 28(9):3260-3263. CAO Danyang, YANG Bingru, LI Guangyuan, et al. K-Medoids clustering algorithm based on CF tree[J]. Application research of computers, 2011, 28(9):3260-3263.
[6] 马箐, 谢娟英. 基于粒计算的K-Medoids聚类算法[J]. 计算机应用, 2012, 32(7):1973-1977. MA Qing, XIE Juanying. New K-Medoids clustering algorithm based on granular computing[J]. Journal of computer applications, 2012, 32(7):1973-1977.
[7] 王常武, 刘小凤, 王宝文, 等. IC-Kmedoids:适用于RNA二级结构预测的聚类算法[J]. 生物医学工程学杂志, 2015, 32(1):99-103. WANG Changwu, LIU Xiaofeng, WANG Baowen, et al. IC-Kmedoids:a clustering algorithm for RNA secondary structure prediction[J]. Journal of biomedical engineering, 2015, 32(1):99-103.
[8] 赵烨, 黄泽君. 蚁群K-Medoids融合的聚类算法[J]. 电子测量与仪器学报, 2012, 26(9):800-804. ZHAO Ye, HUANG Zejun. Clustering algorithm based on fusion of ant colony algorithm and K-Medoids[J]. Journal of electronic measurement and instrument, 2012, 26(9):800-804.
[9] KAUFMAN L, ROUSSEEUW P J. Partitioning around medoids (Program PAM)[M]//KAUFMAN L, ROUSSEEUW P J. Finding groups in Data:An Introduction to Cluster Analysis. Hoboken:John Wiley & Sons, Inc., 2008:68-125.
[10] KAUFMAN L, ROUSSEEUW P J. Clustering large applications (Program CLARA)[M]//KAUFMAN L, ROUSSEEUW P J. Finding Groups in Data:An Introduction to Cluster Analysis. Hoboken:John Wiley & Sons, Inc., 2008:126-163.
[11] 赵庆. 基于Hadoop平台下的Canopy-Kmeans高效算法[J]. 电子科技, 2014, 27(2):29-31. ZHAO Qing. Efficient algorithm of Canopy-Kmeans based on Hadoop platform[J]. Electronic science & technology, 2014, 27(2):29-31.
[12] TENENBAUM J B, DE SILVA V, LANGFORD J C. A global geometric framework for nonlinear dimensionality reduction[J]. Science, 2000, 290(5500):2319-2323.
[13] 孙岩. 湿地植物高光谱特征分析与物种识别模型构建[D]. 北京:清华大学, 2008. SUN Yan. Hyperspectral characters and species identification model for wetland plants[D]. Beijing:Tsinghua University, 2008.
[14] XIANG Shiming, NIE Feiping, ZHANG Changshui. Learning a mahalanobis distance metric for data clustering and classification[J]. Pattern recognition, 2008, 41(12):3600-3612.
[15] KEERTHI S S, SHEVADE S K, BHATTACHARYYA C, et al. A fast iterative nearest point algorithm for support vector machine classifier design[J]. IEEE transactions on neural networks archive, 2000, 11(1):124-136.
[16] MO Jiahui, KIANG M Y, ZOU Peng, et al. A two-stage clustering approach for multi-region segmentation[J]. Expert systems with applications, 2010, 37(10):7120-7131.
[17] 李荟娆. K-means聚类方法的改进及其应用[D]. 哈尔滨:东北农业大学, 2014. LI Huirao. Improved K-means clustering method and its application[D]. Harbin:Northeast Agricultural University, 2014.
[18] 张人. 一种新的动态XML文档聚类方法[D]. 长春:吉林大学, 2012. ZHANG Ren. A novel clustering method for dynamic XML documents[D]. Changchun:Jilin University, 2012.

相似文献/References:

[1]赵春晖,朱志球.二进脊波变换的高光谱遥感图像融合分类[J].哈尔滨工程大学学报,2008,(11):1222.
 ZHAO Chun-hui,ZHU Zhi-qiu.Fusion classification of hyperspectral remote sensing images by dyadic ridgelet transform[J].hebgcdxxb,2008,(09):1222.
[2]赵春晖,崔士玲,赵艮平.改进的多端元高光谱解混算法[J].哈尔滨工程大学学报,2015,(09):1281.[doi:10.11990/jheu.201405042]
 ZHAO Chunhui,CUI Shiling,ZHAO Genping.An improved multi-endmember hyperspectral unmixing algorithm[J].hebgcdxxb,2015,(09):1281.[doi:10.11990/jheu.201405042]
[3]王立国,赵亮,刘丹凤.SVM在高光谱图像处理中的应用综述[J].哈尔滨工程大学学报,2018,39(06):973.[doi:10.11990/jheu.201704074]
 WANG Liguo,ZHAO Liang,LIU Danfeng.A review on the application of SVM in hyperspectral image processing[J].hebgcdxxb,2018,39(09):973.[doi:10.11990/jheu.201704074]

备注/Memo

备注/Memo:
收稿日期:2016-6-28。
基金项目:国家自然科学基金项目(61675051);黑龙江省自然科学基金项目(F201409).
作者简介:王立国(1974-),男,教授,博士生导师.
通讯作者:王立国,E-mail:wangliguo@hrbeu.edu.cn
更新日期/Last Update: 2018-09-12