期刊文献+

高光谱遥感图像DE-self-training半监督分类算法 被引量:8

Semi-supervised Classification Algorithm for Hyperspectral Remote Sensing Image Based on DE-self-training
下载PDF
导出
摘要 提出了一种高光谱遥感图像半监督分类算法DE-self-training。利用少量标记样本作为初始训练集,基于改进的Self-training算法构建初始分类器,对未标记样本进行预测;然后从分类结果中按一定比例随机选取部分样本,连同其类别标记一起加入训练集中,再用扩大的训练集重新训练分类器,并对剩余的未标记样本进行预测。如此迭代地进行训练-预测-挑选样本扩大训练集过程。同时,在迭代训练过程中,运用基于最近邻域规则的数据剪辑策略对扩大训练集时产生的误标记样本进行过滤,以保证训练集的质量,不断迭代地训练出更精确的分类器,最终使所有未标记样本都获得类别标记。以AVIRIS Indian Pines和Hyperion EO-1 Botswana作为实验数据对DE-self-training算法进行测试,并与基于支持向量机的分类结果作比对。实验表明,DE-self-training算法可以在标记样本数量有限条件下,充分挖掘未标记样本的有用信息,使总体分类精度和Kappa系数都有不同程度的提高。 A semi-supervised classification algorithm named DE-self-training for hyperspectral remote sensing images was proposed. Firstly, taking a few labeled samples as initial training set, the initial classification model was constructed by using improved Self-training algorithm to classify unlabeled samples. Then, partial samples and corresponding labels were selected randomly as a proportion from classification results into training set, and the augmented training set was used to retrain the model to classify the unlabeled samples. Then, the algorithm continued the process of training-classifying-picking out samples to augment training set iteratively. During this process, in order to ensure the training set' s quality and the correct labeling of new increased samples, the algorithm edited and purified mislabeled samples by using data editing strategy based on the nearest neighbor rule. Finally, the proposed algorithm trained classification model iteratively to get a more accurate result until the unlabeled samples set was empty. In the experiments, AVIRIS Indian Pines and Hyperion EO - 1 Botswana data were used to test the algorithm. According to the comparison with SVM classification results, accuracy and Kappa coefficients by utilizing unlabeled samples the DE-self-training algorithm can get higher information under limited labeled samples.
出处 《农业机械学报》 EI CAS CSCD 北大核心 2015年第5期239-244,共6页 Transactions of the Chinese Society for Agricultural Machinery
基金 国家自然科学基金资助项目(41171269) 江苏省高校自然科学研究面上项目(14KJB170010) 环保公益性行业科研专项项目(201309037) 江苏高校优势学科建设工程资助项目(164320H101) 地球系统科学数据共享平台资助项目(2005DKA32300) 江苏省普通高校研究生科研创新计划资助项目(1812000002A403)
关键词 高光谱遥感图像 半监督分类 数据剪辑 Hyperspectral remote sensing image Semi-supervised classification Data editing
  • 相关文献

参考文献18

  • 1孙伟伟.基于流形学习的高光谱影像降维理论与方法研究[J].测绘学报,2014,43(4):439-439. 被引量:8
  • 2Shahshahani B, Landgrebe D. 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.
  • 3Zhu X. Semi-supervised learning literature survey [ R]. Technical Report 1530, Department of Computer Sciences, University ofWisconsin, Madison, 2005.
  • 4Blum A, Mitchell T. Combining labeled and unlabeled data with co-training [ C ]//Peter B, Yishay M. Proceedings of the 11 th Annual Conference on Computational Learning theory. Madison : ACM Press, 1998 : 92 - 100.
  • 5Brefeld U, Grtner T, Scheffer T, et al. Efficient co-regularised least squares regression [ C ] //Proceedings of the 23rd International Conference on Machine Learning, 2006:137 -144.
  • 6Blum A, Lafferty J, Rwebangira M, et al. Semi-supervised learning using randomized mincuts[ C]//Brndley C E. Proceedings of the 21st International Conference on Machine Learning (ICML 2004). Banff: ACM Press, 2004: 934- 947.
  • 7Nigam K, Ghani R. Analyzing the effectiveness and applicability of co-training [ C ] //Proceedings of the Ninth International Conference on Information and Knowledge Management, 2000:86 -93.
  • 8Miller D J, Browning J. A mixture model and EM-based algorithm for class discovery, robust classification, and outlier rejection in mixed labeled/unlabeled data sets [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25( 11 ) :1468 - 1483.
  • 9刘小芳,何彬彬,李小文.基于半监督核模糊c-均值算法的北京一号小卫星多光谱图像分类[J].测绘学报,2011,40(3):301-306. 被引量:9
  • 10朱佳丽,李士进,万定生,冯钧.基于特征选择和半监督学习的遥感图像检索[J].中国图象图形学报,2011,16(8):1474-1482. 被引量:8

二级参考文献38

共引文献22

同被引文献157

  • 1姚付启,张振华,杨润亚,孙金伟,崔素芳.基于红边参数的植被叶绿素含量高光谱估算模型[J].农业工程学报,2009,25(S2):123-129. 被引量:76
  • 2刘春红,赵春晖,张凌雁.一种新的高光谱遥感图像降维方法[J].中国图象图形学报(A辑),2005,10(2):218-222. 被引量:81
  • 3苏红军,杜培军.高光谱数据特征选择与特征提取研究[J].遥感技术与应用,2006,21(4):288-293. 被引量:44
  • 4许迪,龚时宏.大型灌区节水改造技术支撑体系及研究重点[J].水利学报,2007,38(7):806-811. 被引量:45
  • 5郭元裕.农田水利学[M].北京:中国水利水电出版社,1997..
  • 6中华人民共和国水利部.中国水资源公报2013[M].北京:中国水利水电出版社,2014.
  • 7KUO B C, LANDGREBE D A. A Robust Classification Procedure Based on Mixture Classifiers and Nonparametric Weighted Feature Extraction[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(11): 2486-2494.
  • 8LEE C, LANDGREBE D A. Feature Extraction Based on Decision Boundaries[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1993, 15(4): 388-400.
  • 9ZHOU Zhihua, ZHAN Dechuan, YANG Qiang. Semi-supervised Learning with Very Few Labeled Training Examples[C]//Proceedings of the National Conference on Artificial Intelligence. Cambridge, MA London: [s.n.], 2007, 22(1): 675-680.
  • 10CHAWLA N V, KARAKOULAS G. Learning from Labeled and Unlabeled Data: An Empirical Study across Techniques and Domains[J]. Journal of Artificial Intelligence Research, 2005, 23(1): 331-366.

引证文献8

二级引证文献68

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部