期刊
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
卷 59, 期 1, 页码 508-521出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TGRS.2020.2997863
关键词
Feature extraction; Hyperspectral imaging; Neural networks; Generative adversarial networks; Gallium nitride; Adversarial learning; classification; domain adaptation; remote sensing
类别
资金
- National Natural Science Foundations of China [61771437, 61102104, 91442201]
- Open Research Fund of Key Laboratory of Spectral Imaging Technology, Chinese Academy of Sciences [LSIT201702D]
This study investigates class-wise adversarial adaptation networks for the classification of hyperspectral remote sensing images. By adversarial learning between feature extractor and multiple domain discriminators, domain-invariant features are generated, and a probability-prediction MMD method is introduced to improve feature-alignment performance. The proposed CDA network can achieve unsupervised classification of target images and has demonstrated efficiency in experiments using Hyperion and AVIRIS hyperspectral data.
Class-wise adversarial adaptation networks are investigated for the classification of hyperspectral remote sensing images in this article. By adversarial learning between the feature extractor and the multiple domain discriminators, domain-invariant features are generated. Moreover, a probability-prediction-based maximum mean discrepancy (MMD) method is introduced to the adversarial adaptation network to achieve a superior feature-alignment performance. The class-wise adversarial adaptation in conjunction with the class-wise probability MMD is denoted as the class-wise distribution adaptation (CDA) network. The proposed CDA does not require labeled information in the target domain and can achieve an unsupervised classification of the target image. The experimental results using the Hyperion and Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) hyperspectral data demonstrated its efficiency.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据