Citation: Zheng, H.; Cao, Y.; Sun, M.;
Guo, G.; Meng, J.; Guo, X.; Jiang, Y.
Mixed Structure with 3D Multi-
Shortcut-Link Networks for
Hyperspectral Image Classification.
Remote Sens. 2022, 14, 1230. https://
doi.org/10.3390/rs14051230
Academic Editors: Yangquan Chen,
Subhas Mukhopadhyay, Nunzio
Cennamo, M. Jamal Deen,
Junseop Lee and Simone Morais
Received: 28 January 2022
Accepted: 28 February 2022
Published: 2 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Mixed Structure with 3D Multi-Shortcut-Link Networks
for Hyperspectral Image Classification
Hui Zheng
1,2
, Yizhi Cao
1,
*, Min Sun
2
, Guihai Guo
1
, Junzhen Meng
1
, Xinwei Guo
1
and Yanchi Jiang
3
1
Ural Institute, North China University of Water Resource and Electric Power, Zhengzhou 450045, China;
zhenghui@ncwu.edu.cn (H.Z.); ggh@ncwu.edu.cn (G.G.); mengjunzhen@ncwu.edu.cn (J.M.);
guoxinwei@ncwu.edu.cn (X.G.)
2
School of Earth and Space Sciences, Peking University, Beijing 100871, China; sunmin@pku.edu.cn
3
Department of Applied Chemistry, Chubu University, 1200 Matsumoto-cho, Kasugai 487-8501, Japan;
jiangyc@isc.chubu.ac.jp
* Correspondence: caoyizhi@ncwu.edu.cn
Abstract:
A hyperspectral image classification method based on a mixed structure with a 3D multi-
shortcut-link network (MSLN) was proposed for the features of few labeled samples, excess noise,
and heterogeneous homogeneity of features in hyperspectral images. First, the spatial–spectral joint
features of hyperspectral cube data were extracted through 3D convolution operation; then, the deep
network was constructed and the 3D MSLN mixed structure was used to fuse shallow representational
features and deep abstract features, while the hybrid activation function was utilized to ensure the
integrity of nonlinear data. Finally, the global self-adaptive average pooling and L-softmax classifier
were introduced to implement the terrain classification of hyperspectral images. The mixed structure
proposed in this study could extract multi-channel features with a vast receptive field and reduce the
continuous decay of shallow features while improving the utilization of representational features and
enhancing the expressiveness of the deep network. The use of the dropout mechanism and L-softmax
classifier endowed the learned features with a better generalization property and intraclass cohesion
and interclass separation properties. Through experimental comparative analysis of six groups
of datasets, the results showed that this method, compared with the existing deep-learning-based
hyperspectral image classification methods, could satisfactorily address the issues of degeneration
of the deep network and “the same object with distinct spectra, and distinct objects with the same
spectrum.” It could also effectively improve the terrain classification accuracy of hyperspectral images,
as evinced by the overall classification accuracies of all classes of terrain objects in the six groups of
datasets: 97.698%, 98.851%, 99.54%, 97.961%, 97.698%, and 99.138%.
Keywords:
hyperspectral image classification; 3D multi-shortcut-link networks; large softmax;
SELU; PReLU
1. Introduction
Hyperspectral images (HSIs) contain rich spatial and spectral information [
1
,
2
] and
are widely applied in the fields of precision agriculture [
3
], urban planning [
4
], national
defense construction [
5
], and mineral exploitation [
6
], among other fields. It has been very
successful in allowing active users to participate in collecting, updating, and sharing the
massive amounts of data that reflect human activities and social attributes [7–10].
The terrain classification of hyperspectral images is a fundamental problem for various
applications, where the classification aims to assign a label with unique class attributes
to each pixel in the image based on the sample features of HSI. However, an HSI is high-
dimensional with few labeled samples, images between wavebands have a high correlation,
and terrain objects with heterogeneous structures that may be homogeneous present the
terrain classification of HSIs with huge challenges.
Remote Sens. 2022, 14, 1230. https://doi.org/10.3390/rs14051230 https://www.mdpi.com/journal/remotesensing
