Hyperspectral Image Classification With Canonical Correlation Forests
Hyperspectral Image Classification With Canonical Correlation Forests
复制标题
DOI：
10.1109/tgrs.2016.2607755
复制
发表时间：
2017
影响因子：
8.2
通讯作者：
J. Xia;N. Yokoya;A. Iwasaki
中科院分区：
文献类型：
--
作者：
J. Xia;N. Yokoya;A. Iwasaki
多分类器系统或集成学习是提供高光谱遥感图像准确分类结果的有效工具。两个著名的集成学习分类器的高光谱数据是随机森林（RF）和旋转森林（RoF）。在本文中，我们提出了一种新的决策树（DT）集成方法，即典型相关森林（CCF）。更具体地说，几个单独的典型相关树（CCT），是二进制DT，使用典型相关分量的超平面分裂，被用来构建CCF。此外，我们在CCF中采用投影引导技术，在投影空间中保留完整的光谱带进行分裂选择。上述技术允许CCF提高成员分类器的准确性和集合内的多样性。此外，CCF扩展到频谱空间框架，包括马尔可夫随机场，扩展的多属性配置文件（EMAPs），和独立分量分析和滚动制导滤波器（E-ICA-RGF）的合奏。在6个高光谱数据集上的实验结果表明，与RF和RoF方法相比，CCF方法在精度和计算复杂度方面具有较好的效果，证明CCF方法是一种很有前途的高光谱图像分类方法，不仅可以利用光谱信息，而且可以在光谱-空间框架内进行分类。
Multiple classifier systems or ensemble learning is an effective tool for providing accurate classification results of hyperspectral remote sensing images. Two well-known ensemble learning classifiers for hyperspectral data are random forest (RF) and rotation forest (RoF). In this paper, we proposed to use a novel decision tree (DT) ensemble method, namely, canonical correlation forest (CCF). More specifically, several individual canonical correlation trees (CCTs) that are binary DTs, which use canonical correlation components for the hyperplane splitting, are used to construct the CCF. Additionally, we adopt the projection bootstrap technique in CCF, in which the full spectral bands are retained for split selection in the projected space. The techniques aforementioned allow the CCF to improve the accuracy of member classifiers and diversity within the ensemble. Furthermore, the CCF is extended to the spectral-spatial frameworks that incorporate Markov random fields, extended multiattribute profiles (EMAPs), and the ensemble of independent component analysis and rolling guidance filter (E-ICA-RGF). Experimental results on six hyperspectral data sets are used to indicate the comparative effectiveness of the proposed method, in terms of accuracy and computational complexity, compared with RF and RoF, and it turns out that CCF is a promising approach for hyperspectral image classification not only with spectral information but also in the spectral-spatial frameworks.