Dual U-Nets autoencoders for unsupervised hyperspectral image super-resolution

Jiaxin Li; Ke Zheng; Li Ni; Lianru Gao

doi:10.1117/12.3010344

15 November 2023 Dual U-Nets autoencoders for unsupervised hyperspectral image super-resolution

Jiaxin Li, Ke Zheng, Li Ni, Lianru Gao

Proceedings Volume 12815, International Conference on Remote Sensing, Mapping, and Geographic Systems (RSMG 2023); 128150J (2023) https://doi.org/10.1117/12.3010344
Event: International Conference on Remote Sensing, Mapping, and Geographic Systems (RSMG 2023), 2023, Kaifeng, China

Abstract

Super-resolution is a burgeoning technique to enhance the spatial resolution of hyperspectral images (HSI) by fusing them with an auxiliary high-resolution multispectral image (MSI). However, most deep learning-based methods are based on supervised paradigm, which rely on large number of training triplets. Considering their deficiency, we propose a dual U-Nets autoencoders network inspired by the theory of spectral mixing, which purely operates on one pair of HSI-MSI for network optimization. Specifically, two coupled autoencoders are deployed as the backbone of our method, aiming to extract latent endmember and abundance of target image. To enrich the representation ability and enhance the interaction of two data, we design a novel dual U-Nets architecture for the encoder part by a parameter-shared strategy. Experiments in Chikusei dataset demonstrate the effectiveness of out method when compared with other state-of-the-art methods.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Jiaxin Li, Ke Zheng, Li Ni, and Lianru Gao "Dual U-Nets autoencoders for unsupervised hyperspectral image super-resolution", Proc. SPIE 12815, International Conference on Remote Sensing, Mapping, and Geographic Systems (RSMG 2023), 128150J (15 November 2023); https://doi.org/10.1117/12.3010344

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Hyperspectral imaging

Super resolution

Design and modelling

Feature extraction

Image enhancement

Computer architecture

Convolution

Show All Keywords

Keywords/Phrases

Search In:

Publication Years