Representation Space-Based Discriminative Graph Construction for Semisupervised Hyperspectral Image Classification

• Published in: IEEE signal processing letters 2018-01, Vol.25 (1), p.35-39
• Main Authors: Shao, Yuanjie, Sang, Nong, Gao, Changxin
• Format: Article
• Language: English
• Subjects: Graph; hyperspectral image (HSI) classification; Hyperspectral imaging; Laplace equations; Linear programming; Noise measurement; Optimization; Probabilistic logic; representation space; Semisupervised learning; semisupervised learning (SSL)
• ISSN: 1070-9908; 1558-2361
• DOI: 10.1109/LSP.2017.2764103

Graph-based semisupervised learning methods have been successfully applied in hyperspectral image (HSI) classification with limited labeled samples. The critical step of graph-based methods is to learn a similarity graph, and numerous graph construction methods have been developed in recent years. However, existing approaches usually return a similarity matrix from the raw data space. In this letter, we propose a representation space-based discriminative graph for semisupervised HSI classification, which can learn the representations of samples and the similarity matrix of representations simultaneously. Moreover, we explicitly incorporate the probabilistic class relationship between sample and class, which can be estimated by the partial label information, into the above model to further boost the discriminability of graph. The experimental results on Hyperion and AVIRIS hyperspectral data demonstrate the effectiveness of the proposed approach.