[1]王立国,杨月霜,刘丹凤.基于改进三重训练算法的高光谱图像半监督分类[J].哈尔滨工程大学学报,2016,37(06):849-854.[doi:10.11990/jheu.201505078]
 WANG Liguo,YANG Yueshuang,LIU Danfeng.Semi-supervised classification for hyperspectral image based on improved tri-training method[J].hebgcdxxb,2016,37(06):849-854.[doi:10.11990/jheu.201505078]
点击复制

基于改进三重训练算法的高光谱图像半监督分类(/HTML)
分享到:

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

卷:
37
期数:
2016年06期
页码:
849-854
栏目:
出版日期:
2016-06-25

文章信息/Info

Title:
Semi-supervised classification for hyperspectral image based on improved tri-training method
作者:
王立国 杨月霜 刘丹凤
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 YANG Yueshuang LIU Danfeng
College of Information and Communications Engineering, Harbin Engineering University, Harbin 150001, China
关键词:
高光谱图像半监督分类三重训练边缘采样差分进化
分类号:
TP75
DOI:
10.11990/jheu.201505078
文献标志码:
A
摘要:
高光谱数据维数高,有标签的样本数量少,给高光谱图像分类带来困难。本文针对传统三重训练(tri-training)算法在初始有标签样本数量较少的情况下分类器间差异性不足的问题提出了一种基于改进三重训练算法的半监督分类框架。该方法首先通过边缘采样策略(margin Sampling,MS)选取最富含信息量的无标签样本,然后在训练每个分类器之前通过差分进化算法(differential evolution,DE)利用所选取的无标签样本产生新的样本。这些新产生的样本将被标记并且加入训练样本集来帮助初始化分类器。实验结果表明,该方法不仅能够有效地利用无标签样本,而且在有标签数据很少的情况下能够有效地提高分类精度。

参考文献/References:

[1] WANG Liguo, JIA Xiuping. Integration of soft and hard classifications using extended support vector machines[J]. IEEE geoscience and remote sensing letters, 2009, 6(3):543-547.
[2] SHAHSHAHANI B M, LANDGREBE D A. The effect of unlabeled samples in reducing the small sample size problem and mitigating the hughes phenomenon[J]. IEEE transactions on geoscience and remote sensing, 1994, 32(5):1087-1095.
[3] ZHU Xiaojin. Semi-supervised learning literature survey[D]. Madison:University of Wisconsin-Madison, 2008.
[4] BARALDI A, BRUZZONE L, BLONDA P. A multiscale expectation maximization semisupervised classifier suitable for badly posed image classification[J]. IEEE transactions on image processing, 2006, 15(8):2208-2225.
[5] JOACHIMS T. Transductive inference for text classification using support vector machines[C]//Proceedings of the 16th International Conference on Machine Learning. Bled, Slovenia, 1999:200-209.
[6] CHI Mingmin, BRUZZONE L. Classification of hyperspectral data by continuation semi-supervised SVM[C]//Proceedings of the 2007 IEEE International Geoscience and Remote Sensing Symposium. Barcelona, 2007:3794-3797.
[7] BLUM A, CHAWLA S. Learning from labeled and unlabeled data using graph mincuts[C]//Proceedings of the 18th International Conference on Machine Learning. Williamston, 2001:19-26.
[8] BLUM A, MITCHELL T. Combining labeled and unlabeled data with co-training[C]//Proceedings of the 11th Annual Conference on Computational Learning Theory. Madison, 1998:92-100.
[9] GOLDMAN S, ZHOU Yan. Enhancing supervised learning with unlabeled data[C]//Proceedings of the 17th international conference on machine learning. San Francisco, CA, 2000:327-334.
[10] ZHOU Zhihua, LI Ming. Tri-training:Exploiting unlabeled data using three classifiers[J]. IEEE transactions on knowledge and data engineering, 2005, 17(11):1529-1541.
[11] ZHANG Youmin, YU Zhengtao, LIU Li, et al. Semi-supervised expert metadata extraction based on co-training style[C]//Proceedings of the 9th international conference on fuzzy systems and knowledge discovery. Chongqing, 2012:1344-1347.
[12] LI Kunlun, ZHANG Wei, MA Xiaotao, et al. A novel semisupervised svm based on tri-training[C]//Proceedings of the 2nd International Symposium on Intelligent Information Technology Application. Shanghai, China, 2008:47-51.
[13] BREIMAN L. Bagging predictors[J]. Machine learning, 1996, 24(2):123-140.
[14] LI Ming, ZHOU Zhihua. Improve computer-aided diagnosis with machine learning techniques using undiagnosed samples[J]. IEEE transactions on systems, man, and cybernetics, part A:systems and humans, 2007, 37(6):1088-1098.
[15] TRIGUERO I, GARCIA S, HERRERA F. SEG-SSC:a framework based on synthetic examples generation for self-labeled semi-supervised classification[J]. IEEE transactions on cybernetics, 2015, 45(4):622-634.
[16] PRICE K V, STORN R M, LAMPINEN J A. Differential evolution:a practical approach to global optimization[M]. Berlin Heidelberg:Springer, 2005:292.
[17] MACKAY D J C. Information-based objective functions for active data selection[J]. Neural computation, 1992, 4(4):590-604.
[18] SCHOHN G, COHN D. Less is more:Active learning with support vectors machines[C]//Proceedings of the 17th international conference on machine learning. Stanford, CA, 2000:839-846.
[19] CAMPBELL C, CRISTIANINI N, SMOLA A. Query learning with large margin classifiers[C]//Proceedings of the 17th international conference on machine learning. Stanford, CA, 2000:111-118.
[20] NGUYEN H T, SMEULDERS A. Active learning using pre-clustering[C]/Proceedings of the 21th international conference on machine learning. Banff, AB, Canada, 2004:79.
[21] FREUND Y, SEUNG H, SHAMIR E, et al. Selective sampling using the query by committee algorithm[J]. Machine learning, 1997, 28(2/3):133-168.

相似文献/References:

[1]赵春晖,胡春梅,石红.采用选择性分段PCA算法的高光谱图像异常检测[J].哈尔滨工程大学学报,2010,(11):0.
 College of Information and Communication Engineering,Harbin Engineering University,Harbin,et al.Anomaly detection algorithm for hyperspectral image by using selective section principal component analysis[J].hebgcdxxb,2010,(06):0.
[2]赵春晖,胡春梅,石红.采用选择性分段PCA算法的高光谱图像异常检测[J].哈尔滨工程大学学报,2011,(01):109.[doi:doi:10.3969/j.issn.1006-7043.2011.01.020]
 ZHAO Chunhui,HU Chunmei,SHI Hong.Anomaly detection for a hyperspectral image by using a selective section principal component analysis algorithm[J].hebgcdxxb,2011,(06):109.[doi:doi:10.3969/j.issn.1006-7043.2011.01.020]
[3]刘振林,谷延锋,张晔.一种用于高光谱图像特征提取的子空间核方法[J].哈尔滨工程大学学报,2014,(02):238.[doi:10.3969/j.issn.10067043.201309025]
 LIU Zhenlin,GU Yanfeng,ZHANG Ye.A subspace kernel learning method for feature extraction of the hyperspectral image[J].hebgcdxxb,2014,(06):238.[doi:10.3969/j.issn.10067043.201309025]
[4]刘丹凤,王立国,赵亮.高光谱图像的同步彩色动态显示[J].哈尔滨工程大学学报,2014,(06):760.[doi:10.3969/j.issn.10067043.201306008]
 LIU Danfeng,WANG Liguo,ZHAO Liang.Dynamic display of the hyperspectral image synchronized colors[J].hebgcdxxb,2014,(06):760.[doi:10.3969/j.issn.10067043.201306008]
[5]杨京辉,王立国,钱晋希.基于相关向量机的高光谱图像解混方法[J].哈尔滨工程大学学报,2015,(02):267.[doi:10.3969/j.issn.1006-7043.201311016]
 YANG Jinghui,WANG Liguo,QIAN Jinxi.An unmixing algorithm based on the relevance vector machine for hyperspectral imagery[J].hebgcdxxb,2015,(06):267.[doi:10.3969/j.issn.1006-7043.201311016]
[6]赵春晖,靖晓昊,李威.基于StOMP稀疏方法的高光谱图像目标检测[J].哈尔滨工程大学学报,2015,(07):992.[doi:10.3969/j.issn.1006-7043.201404087]
 ZHAO Chunhui,JING Xiaohao,LI Wei.Hyperspectral image target detection algorithm based on StOMP sparse representation[J].hebgcdxxb,2015,(06):992.[doi:10.3969/j.issn.1006-7043.201404087]
[7]赵春晖,王佳,王玉磊.采用背景抑制和自适应阈值分割的高光谱异常目标检测[J].哈尔滨工程大学学报,2016,37(02):278.[doi:10.11990/jheu.201409035]
 ZHAO Chunhui,WANG Jia,WANG Yulei.Hyperspectral anomaly detection based on background suppression and adaptive threshold segmentation[J].hebgcdxxb,2016,37(06):278.[doi:10.11990/jheu.201409035]
[8]王立国,宛宇美,路婷婷,等.结合经验模态分解和Gabor滤波的高光谱图像分类[J].哈尔滨工程大学学报,2016,37(02):284.[doi:10.11990/jheu.201411032]
 WANG Liguo,WAN Yumei,LU Tingting,et al.Hyperspectral image classification by combining empirical mode decomposition with Gabor filtering[J].hebgcdxxb,2016,37(06):284.[doi:10.11990/jheu.201411032]
[9]赵春晖,王鑫鹏,闫奕名.基于密度背景纯化的高光谱异常检测算法[J].哈尔滨工程大学学报,2016,37(12):1722.[doi:10.11990/jheu.201511073]
 ZHAO Chunhui,WANG Xinpeng,YAN Yiming.Density background refinement-based anomaly detection algorithm for hyperspectral images[J].hebgcdxxb,2016,37(06):1722.[doi:10.11990/jheu.201511073]
[10]成宝芝,赵春晖,张丽丽.子空间稀疏表示高光谱异常检测新算法[J].哈尔滨工程大学学报,2017,38(04):640.[doi:10.11990/jheu.201604006]
 CHENG Baozhi,ZHAO Chunhui,ZHANG Lili.An anomaly detection algorithm for hyperspectral images using subspace sparse representation[J].hebgcdxxb,2017,38(06):640.[doi:10.11990/jheu.201604006]
[11]盛振国,王立国.改进的LLGC高光谱图像半监督分类[J].哈尔滨工程大学学报,2017,38(07):1086.[doi:10.11990/jheu.201605023]
 SHENG Zhenguo,WANG Liguo.Semi-supervised classification for hyperspectral images based on improved learning with the LLGC method[J].hebgcdxxb,2017,38(06):1086.[doi:10.11990/jheu.201605023]
[12]王立国,李阳.融合主动学习的高光谱图像半监督分类[J].哈尔滨工程大学学报,2017,38(08):1322.[doi:10.11990/jheu.201606046]
 WANG Liguo,LI Yang.Semi-supervised classification for hyperspectral image collaborating with active learning algorithm[J].hebgcdxxb,2017,38(06):1322.[doi:10.11990/jheu.201606046]
[13]崔颖,王雪婷,陆忠军,等.改进M-training算法的高光谱图像分类[J].哈尔滨工程大学学报,2018,39(10):1688.[doi:10.11990/jheu.201707022]
 CUI Ying,WANG Xueting,LU Zhongjun,et al.Hyperspectral image classification based on improved M-training algorithm[J].hebgcdxxb,2018,39(06):1688.[doi:10.11990/jheu.201707022]

备注/Memo

备注/Memo:
收稿日期:2015-05-27。
基金项目:国家自然科学基金项目(60802059);教育部博士点新教师基金项目(200802171003);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,wangliguo@hrbeu.edu.cn.
更新日期/Last Update: 2016-07-05