Distribution Alignment and Discriminative Feature Learning for Domain Adaptation in Hyperspectral Image Classification

Abstract

Domain adaptation (DA) aims to use well-labeled source domain to predict the labels of unlabeled or poor-labeled target domain. Most of existing DA methods focus on the use of feature-level or sample-level information. Recent studies have shown that domain discriminative information is also important for classification. To jointly exploit feature-level information and discriminative information, a new DA method called distribution alignment and discriminative feature learning (DADFL) is proposed for hyperspectral image (HSI) classification in this letter. DADFL incorporates category-discriminative information preservation and structured prediction based pseudo-labeling into a unified framework to simultaneously reduce distribution and subspace differences between domains. Experimental results on three hyperspectral DA tasks show that the classification performance of the proposed DADFL is better than that of existing DA methods.