Iterative Training Sample Expansion to Increase and Balance the Accuracy of Land Classification From VHR Imagery
Iterative Training Sample Expansion to Increase and Balance the Accuracy of Land Classification From VHR Imagery
复制标题
迭代训练样本扩展以提高和平衡 VHR 图像土地分类的准确性
DOI：
10.1109/tgrs.2020.2996064
复制
发表时间：
2020
影响因子：
8.2
通讯作者：
Giles M. Foody
中科院分区：
文献类型：
--
作者：
ZhiYong Lv;GuangFei Li;ZheNong Jin;Jon Atli Benediktsson;Giles M. Foody
众所周知，不平衡的训练集会产生用于监督分类的次优映射。因此，绘制土地覆盖图的一个挑战是获取训练数据，该数据将允许以较高的整体精度进行分类，其中每个类别也映射到类似用户的精度。为了解决这个问题，我们将局部自适应区域和盒须图（BP）技术集成到迭代算法中，以扩大当前研究中选定类别的训练样本的大小。所提出算法的主要步骤如下。首先，手动标记每个类别集的非常小的初始训练样本。其次，通过进行局部光谱变化分析，在自适应区域内找到潜在的新训练样本。最后，获取三个新的训练样本以捕获有关类内变异的信息；这些样本位于 BP 的下四分位数、中四分位数和上四分位数。将这些新的训练样本添加到初始训练样本后，重新训练分类，并且迭代地继续该过程直到终止。所提出的方法应用于三幅超高分辨率（VHR）遥感图像，并与一组同源方法进行比较。比较表明，所提出的方法在整体准确性方面产生了最佳结果，并且在平衡用户准确性方面表现出优越性。例如，就整体精度而言，所提出的方法通常比比较方法高 2%-10%，并且通常会产生最平衡的分类。
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 in which each class is also mapped onto similar user’s accuracy. To solve this problem, we integrated local adaptive region and boxand-whisker plot (BP) techniques into an iterative algorithm to expand the size of the training sample for selected classes in the current study. The major steps of the proposed algorithm are as follows. First, a very small initial training sample 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 intra-class variation; these samples lie in the lower, median, and upper quartiles of BP. After adding these new training samples to the initial training sample, 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 overall accuracy and exhibited superiority in balancing user’s accuracy. For example, the proposed approach was typically 2%-10% more accurate than the compared methods in terms of overall accuracy and it generally yielded the most balanced classification.