文献类型： 外文期刊

作者： Cui, Ying ¹ ; Song, Guojiao ¹ ; Wang, Xueting ¹ ; Lu, Zhongjun ² ; Wang, Liguo ¹ ;

作者机构： 1.Harbin Engn Univ, Coll Informat & Commun Engn, Harbin, Heilongjiang, Peoples R China

2.Heilongjiang Acad Agr Sci, Remote Sensing Technol Ctr, Harbin, Heilongjiang, Peoples R China

关键词： hyperspectral image;semisupervised classification;tri-training algorithm;diversification of classifiers;stratified sampling

期刊名称：JOURNAL OF APPLIED REMOTE SENSING （ 影响因子：1.53； 五年影响因子：1.565 ）

ISSN： 1931-3195

年卷期： 2017 年 11 卷

页码：

收录情况： SCI

摘要： Hyperspectral image classification faces a serious challenge due to the high dimension of hyperspectral data and limited labeled samples. Tri-training algorithm is a widely used semi-supervised classification method, but the algorithm lacks significant diversity among the classifiers when the number of initial label samples is limited. A semi-supervised classification method for hyperspectral data based on tri-training is proposed. It combines different classifiers and stratified sampling based on labeled class to increase classifier diversity and avoid classifier performance deterioration. Performance comparison between the proposed algorithm and tri-training algorithm was made through experiments. The proposed algorithm improved the overall accuracy and Kappa coefficient by 1.37% to 6.84% and 0.0096 to 0.0808, respectively, and the results showed that the effectiveness of the algorithm is verified. Moreover, the algorithm can also get better performance when the number of samples is small. (C) 2017 Society of Photo-Optical Instrumentation Engineers (SPIE)