Discriminative Feature Metric Learning in the Affinity Propagation Model for Band Selection in Hyperspectral Images
Discriminative Feature Metric Learning in the Affinity Propagation Model for Band Selection in Hyperspectral Images
复制标题
DOI：
10.3390/rs9080782
复制
发表时间：
2017-07
期刊：
影响因子：
--
通讯作者：
Chen Yang;Yulei Tan;L. Bruzzone;Laijun Lu;Renchu Guan
中科院分区：
文献类型：
--
作者：
Chen Yang;Yulei Tan;L. Bruzzone;Laijun Lu;Renchu Guan
传统的有监督波段选择方法主要考虑通过减少光谱冗余来提高高光谱图像分类精度。一个关键的观察结果是，在HSI中空间上彼此接近的像素可能具有相同的特征，而空间上彼此远离的像素很可能属于不同的类别。在本文中，我们提出了一种新的歧视性的特征度量为基础的亲和传播（DFM-AP）技术，像素之间的光谱和空间关系的构造一种新的类型的歧视性约束。这种区分性约束包括小块和区分性信息，它们被引入BS过程。小块信息允许将光谱上接近的像素和空间上接近的像素分组在一起，而不需要它们的类别标签的明确知识，而判别信息提供重要的可分性信息。提出了一种区分性特征度量（DFM），其区分性约束以最优准则建模，用于识别有效的距离度量学习方法，该方法涉及区分性分量分析（DCA）。在此之后，可以通过基于范例的聚类算法来识别带的代表性子集，该聚类算法也被称为亲和传播过程。实验结果表明，与几种典型的基于类标签和成对约束的BS算法相比，该方法具有更好的性能。提出的DFM-AP通过选择具有低冗余度的高鉴别力波段来提高具有鉴别力约束的分类性能。
Traditional supervised band selection (BS) methods mainly consider reducing the spectral redundancy to improve hyperspectral imagery (HSI) classification with class labels and pairwise constraints. A key observation is that pixels spatially close to each other in HSI have probably the same signature, while pixels further away from each other in the space have a high probability of belonging to different classes. In this paper, we propose a novel discriminative feature metric-based affinity propagation (DFM-AP) technique where the spectral and the spatial relationships among pixels are constructed by a new type of discriminative constraint. This discriminative constraint involves chunklet and discriminative information, which are introduced into the BS process. The chunklet information allows for grouping of spectrally-close and spatially-close pixels together without requiring explicit knowledge of their class labels, while discriminative information provides important separability information. A discriminative feature metric (DFM) is proposed with the discriminative constraints modeled in terms of an optimal criterion for identifying an efficient distance metric learning method, which involves discriminative component analysis (DCA). Following this, the representative subset of bands can be identified by means of an exemplar-based clustering algorithm, which is also known as the process of affinity propagation. Experimental results show that the proposed approach yields a better performance in comparison with several representative class label and pairwise constraint-based BS algorithms. The proposed DFM-AP improves the classification performance with discriminative constraints by selecting highly discriminative bands with low redundancy.