[1]徐凯,崔颖.Stacking Learning在高光谱图像分类中的应用[J].应用科技,2018,45(06):42-46,52.[doi:10.11991/yykj.201712011]
 XU Kai,CUI Ying.Application of Stacking Learning in hyperspectral image classification[J].Applied science and technology,2018,45(06):42-46,52.[doi:10.11991/yykj.201712011]
点击复制

Stacking Learning在高光谱图像分类中的应用(/HTML)
分享到:

《应用科技》[ISSN:1009-671X/CN:23-1191/U]

卷:
第45卷
期数:
2018年06期
页码:
42-46,52
栏目:
现代电子技术
出版日期:
2018-11-05

文章信息/Info

Title:
Application of Stacking Learning in hyperspectral image classification
作者:
徐凯1 崔颖12
1. 哈尔滨工程大学 信息与通信工程学院, 黑龙江 哈尔滨 150001;
2. 黑龙江省农业科学院 遥感技术中心, 黑龙江 哈尔滨 150001
Author(s):
XU Kai1 CUI Ying12
1. College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China;
2. Remote Sensing Technology Center, Heilongjiang Academy of Argicultural Sciences, Harbin 150001, China
关键词:
高光谱图像多分类系统Stacking Learning集成学习交叉验证图像分类特征变换K-Fold
Keywords:
hyperspectral imagemultiple classifier systemStacking Learningensemble learningcross-validationimage classificationfeature transformationK-Fold
分类号:
TP75
DOI:
10.11991/yykj.201712011
文献标志码:
A
摘要:
高光谱图像分类研究中,集成学习能够显著地提高分类效果。但是传统的并行多分类系统对基础分类器有较高要求,即要求差异性及分类均衡。为了解决这一问题,采用Stacking Learning的堆栈式学习方式,首先使用K-Fold和交叉验证的方式进行数据分割和训练,将原始特征进行特征变换,重新构建二级特征。再使用新特征进行对Meta分类器进行训练得到判决分类器,用于样本的最后分类判断。实验结果表明,采用的Stacking Learning方法不依赖基础分类器,且相比较于传统的多分类系统具有更高的精度和良好的稳定性。
Abstract:
In the study of hyperspectral image classification, integrated learning can significantly improve the classification effect. However, traditional parallel multiple classifier system has higher requirements for the basic classifier, namely, the diversity of requirement and the equalization of classification. In order to solve this problem, we use Stacking Learning method. Firstly, K-Fold and cross-validation were used to segment and train the data. The original features were transformed and the secondary features were reconstructed. Further the new feature was used to train the Meta classifier to obtain a decision classifier for the final classification of the sample. Experimental results show that the Stacking Learning method does not rely on basic classifier and has higher accuracy and stability than traditional multiple classifier system.

参考文献/References:

[1] NATH S S, MISHRA G, KAR J, et al. A survey of image classification methods and techniques[C]//Proceedings of 2014 International Conference on Control, Instrumentation, Communication and Computational Technologies. Kanyakumari, India, 2014.
[2] 杜培军, 夏俊士, 薛朝辉, 等. 高光谱遥感影像分类研究进展[J]. 遥感学报, 2016, 20(2):236-256.
[3] XU Lei, KRZYZAK A, SUEN C Y. Methods of combining multiple classifiers and their applications to handwriting recognition[J]. IEEE transactions on systems, man, and cybernetics, 1992, 22(3):418-435.
[4] KUNCHEVA L I. Combining pattern classifiers:methods and algorithms[M]. New York, USA:Wiley-Interscience, 2004.
[5] BENEDIKTSSON J A, CHANUSSOT J, FAUVEL M. Multiple classifier systems in remote sensing:from basics to recent developments[C]//Proceedings of the 7th International Workshop on Multiple Classifier Systems. Prague, Czech Republic, 2007.
[6] ABDALLAH A C B, FRIGUI H, GADER P. Adaptive local fusion with fuzzy integrals[J]. IEEE transactions on fuzzy systems, 2012, 20(5):849-864.
[7] DU Peijun, XIA Junshi, ZHANG Wei, et al. Multiple classifier system for remote sensing image classification:A review[J]. Sensors, 2012, 12(4):4764-4792.
[8] WOLPERT D H. Stacked generalization[J]. Neural networks, 1992, 5(2):241-259.
[9] ESCALERA S, PUERTAS E, RADEVA P, et al. Multi-modal laughter recognition in video conversations[C]//Proceedings of 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Miami, USA, 2009.
[10] RAZMARA M, SARKAR A. Stacking for statistical machine translation[C]//Proceedings of the the 51st Annual Meeting of the Association for Computational Linguistics. Sofia, Bulgaria, 2013.
[11] TZANIS G, BERBERIDIS C, VLAHAVAS I. StackTIS:a stacked generalization approach for effective prediction of translation initiation sites[J]. Computers in biology and medicine, 2012, 42(1):61-69.
[12] XIA Junshi, CHANUSSOT J, DU Peijun, et al. Rotation-based ensemble classifiers for high-dimensional data[M]//IONESCU B, BENOIS-PINEAU J, PIATRIK T, et al. Fusion in Computer Vision. Cham:Springer, 2014:135-160.
[13] CHEN Tianqi, GUESTRIN C. Xgboost:a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, USA, 2016.
[14] RODRIGUEZ J J, KUNCHEVA L I, ALONSO C J. Rotation forest:a new classifier ensemble method[J]. IEEE transactions on pattern analysis and machine intelligence, 2006, 28(10):1619-1630.
[15] XIA Junshi, DU Peijun, HE Xiyan, et al. Hyperspectral remote sensing image classification based on rotation forest[J]. IEEE geoscience and remote sensing letters, 2014, 11(1):239-243.

相似文献/References:

[1]张文升,王立国,孟凡旺.基于嵌套窗口的高光谱图像目标检测[J].应用科技,2009,36(05):12.[doi:10.3969/j.issn.1009-671X.2009.05.004]
 ZHANG Wen-sheng,WANG Li-guo,MENG Fan-wang.Nested spatial window based target detection for hyperspectral images[J].Applied science and technology,2009,36(06):12.[doi:10.3969/j.issn.1009-671X.2009.05.004]
[2]赵春晖,王楠楠.基于背景抑制及顶点成分分析的[J].应用科技,2009,36(09):11.[doi:oi:10.3969/j.issn.1009-671X.2009.09.003]
 ZHAO Chun hui,WANG Nan nan.Anomaly detection of hyperspectral imagery based on background restrain and VCA[J].Applied science and technology,2009,36(06):11.[doi:oi:10.3969/j.issn.1009-671X.2009.09.003]
[3]王立国,邓禄群,张晶.改进的SGA端元选择的快速方法[J].应用科技,2010,37(04):19.[doi:1009-671X (2010) 04-0019-04]
 WANG Li-guo,DENG Lu-qun,ZHANG Jing.A fast endmember selection method based on simplex growing algorithm[J].Applied science and technology,2010,37(06):19.[doi:1009-671X (2010) 04-0019-04]
[4]王立国,赵妍,王群明.基于POCS的高光谱图像超分辨率方法[J].应用科技,2010,37(10):26.[doi:10.3969/j.issn.1009-671X.2010.10.007]
 WANG Li-guo,ZHAO Yan,WANG Qun-ming.POCS based super-resolution method for hyperspectral imagery[J].Applied science and technology,2010,37(06):26.[doi:10.3969/j.issn.1009-671X.2010.10.007]
[5]杨桄1,张俭峰1,赵波2,等.基于 PCA 和 KRX 算法的高光谱异常检测[J].应用科技,2014,41(05):11.[doi:10.3969 / j.issn.1009⁃671X.201312002]
 ,,et al.Anomaly detection based on PCA and KRX in hyperspectral images[J].Applied science and technology,2014,41(06):11.[doi:10.3969 / j.issn.1009⁃671X.201312002]
[6]王立国,杜心平.K均值聚类和孪生支持向量机相结合的高光谱图像半监督分类[J].应用科技,2017,44(03):12.[doi:10.11991/yykj.201606010]
 WANG Liguo,DU Xinping.Semi-supervised classification of hyperspectral images applying the combination of K-mean clustering and twin support vector machine[J].Applied science and technology,2017,44(06):12.[doi:10.11991/yykj.201606010]
[7]王霖郁,刘一博.基于光谱角匹配加权的高光谱图像异常检测[J].应用科技,2017,44(06):20.[doi:10.11991/yykj.201610010]
 WANG Linyu,LIU Yibo.Anomaly detection for hyperspectral image based on weighted spectral angle match[J].Applied science and technology,2017,44(06):20.[doi:10.11991/yykj.201610010]
[8]崔颖,王雪婷,刘述彬,等.基于改进SPA算法的高效降维方法[J].应用科技,2018,45(05):51.[doi:10.11991/yykj.201712002]
 CUI Ying,WANG Xueting,LIU Shubin,et al.An efficient reduction method based on the improved successive projection algorithm[J].Applied science and technology,2018,45(06):51.[doi:10.11991/yykj.201712002]

备注/Memo

备注/Memo:
收稿日期:2017-12-20。
基金项目:国家自然科学基金项目(61675051)
作者简介:徐凯(1992-),男,硕士研究生;崔颖(1979-),女,副教授,博士
通讯作者:崔颖,E-mail:cuiying@hrbeu.edu.cn
更新日期/Last Update: 2018-11-02