期刊
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
卷 59, 期 1, 页码 139-150出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TGRS.2020.2996064
关键词
Training; Sensors; Hyperspectral imaging; Iterative methods; Feature extraction; Land cover classification; training sample collection; very high-resolution remote-sensing image
类别
资金
- National Natural Science Foundation of China [61701396, 61801380]
- Natural Science Foundation of Shaan Xi Province [2018JQ4009]
The proposed approach integrates local adaptive region and box-and-whisker plot techniques into an iterative algorithm to expand the size of the training sample for selected classes. Applied to three very high-resolution remote-sensing images, it outperformed a set of cognate methods in terms of overall accuracy and balancing users accuracy.
Imbalanced training sets are known to produce suboptimal maps for supervised classification. Therefore, one challenge in mapping land cover is acquiring training data that will allow classification with high overall accuracy (OA) in which each class is also mapped onto similar users accuracy. To solve this problem, we integrated local adaptive region and box-and-whisker plot (BP) techniques into an iterative algorithm to expand the size of the training sample for selected classes in this article. The major steps of the proposed algorithm are as follows. First, a very small initial training sample (ITS) for each class set is labeled manually. Second, potential new training samples are found within an adaptive region by conducting local spectral variation analysis. Lastly, three new training samples are acquired to capture information regarding intraclass variation; these samples lie in the lower, median, and upper quartiles of BP. After adding these new training samples to the ITS, classification is retrained and the process is continued iteratively until termination. The proposed approach was applied to three very high-resolution (VHR) remote-sensing images and compared with a set of cognate methods. The comparison demonstrated that the proposed approach produced the best result in terms of OA and exhibited superiority in balancing users accuracy. For example, the proposed approach was typically 2-10 more accurate than the compared methods in terms of OA and it generally yielded the most balanced classification.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据