Extended morphological profiles using auto-associative neural networks for hyperspectral data classification
Extended morphological profiles using auto-associative neural networks for hyperspectral data classification
复制标题
DOI：
10.1109/whispers.2011.6080867
复制
发表时间：
2011-06
期刊：
影响因子：
--
通讯作者：
G. Licciardi;P. Marpu;J. Benediktsson;J. Chanussot
中科院分区：
文献类型：
--
作者：
G. Licciardi;P. Marpu;J. Benediktsson;J. Chanussot
最近，形态剖面被认为是融合光谱和空间信息以产生更好的分类结果的良好工具。一般来说，配置文件是使用主成分分析 (PCA) 导出的特征构建的。自关联神经网络（AANN）可以看作是非线性PCA的实现，用于高光谱数据的无监督特征缩减。在本文中，我们研究了使用 AANN 派生的特征为高光谱数据分类构建扩展形态剖面的适用性。
Recently, morphological profiles have be observed as good tools to fuse spectral and spatial information to produce better classification results. In general, the profiles are built with the features derived using the principal component analysis (PCA). Auto-associative neural network (AANN), which can be seen as an implementation of non-linear PCA is used for unsupervised feature reduction of hyperspectral data. In this paper, we investigate the suitability of the features derived using AANN to build extended morphological profiles for hyperspectral data classification.