Citation: Yang, R.; Kan, J.
Classification of Tree Species in
Different Seasons and Regions Based
on Leaf Hyperspectral Images.
Remote Sens. 2022, 14, 1524.
https://doi.org/10.3390/rs14061524
Academic Editors: Yangquan Chen,
Francois Girard,
Subhas Mukhopadhyay,
Nunzio Cennamo, M. Jamal Deen,
Junseop Lee and Simone Morais
Received: 19 January 2022
Accepted: 19 March 2022
Published: 21 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
Classification of Tree Species in Different Seasons and Regions
Based on Leaf Hyperspectral Images
Rongchao Yang
1,2
and Jiangming Kan
1,2,
*
1
School of Technology, Beijing Forestry University, Beijing 100083, China; yangrongchao@caas.cn
2
Key Laboratory of State Forestry Administration on Forestry Equipment and Automation, Beijing 100083, China
* Correspondence: kanjm@bjfu.edu.cn; Tel.: +86-010-62336137-706
Abstract:
This paper aims to establish a tree species identification model suitable for different
seasons and regions based on leaf hyperspectral images, and to mine a more effective hyperspectral
identification algorithm. Firstly, the reflectance spectra of leaves in different seasons and regions were
analyzed. Then, to solve the problem that 0-element in sparse random (SR) coding matrices affects
the classification performance of error-correcting output codes (ECOC), two versions of supervision-
mechanism-based ECOC algorithms, namely SM-ECOC-V1 and SM-ECOC-V2, were proposed in
this paper. In addition, the performance of the proposed algorithms was compared with that of
six traditional algorithms based on all bands and feature bands. The experiment results show that
seasonal and regional changes have an effect on the reflectance spectra of leaves, especially in the
near-infrared region of 760–1000 nm. When the spectral information of different seasons and different
regions is added into the identification model, tree species can be effectively classified. SM-ECOC-V2
achieves the best classification performance based on both all bands and feature bands. Furthermore,
both SM-ECOC-V1 and SM-ECOC-V2 outperform the ECOC method under SR coding strategy,
indicating the proposed methods can effectively avoid the influence of 0-element in SR coding matrix
on classification performance.
Keywords:
tree species identification; leaf hyperspectral images; seasonal variations; regional variations;
error-correcting output codes
1. Introduction
Forest resources constitute the main body of the terrestrial ecosystem and are the basis
of construction for forestry and ecological environment. They not only provide abundant
material resources for human survival and development but also play an extremely impor-
tant role in the sustainable development of economy, environment and society [1,2]. They
can provide valuable information for estimating the economic value of forests and studying
forest ecosystems to figure out the composition of forest tree species [
3
]. Therefore, the
accurate identification of forest tree species is of great significance to the rational planning,
utilization and protection of forest resources.
In recent years, using hyperspectral remote sensing technology to identify forest
tree species has become a hot spot in forestry remote sensing research. Hyperspectral
imaging technology is an organic combination of imaging technology and spectroscopy
technology which can obtain tens to hundreds of spectral bands for each pixel and reflect
the subtle differences between the reflectance spectra of different ground objects, thus
greatly improving the identification ability of ground objects [
4
,
5
]. According to the
different acquisition methods of hyperspectral data, it can be divided into satellite-borne
hyperspectral data, airborne hyperspectral data and non-imaging hyperspectral data.
Satellite-borne hyperspectral remote sensing technology is convenient to realize the
identification of forest tree species on a large scale. Many researchers have carried out
research on forest tree species identification based on satellite-borne hyperspectral images
Remote Sens. 2022, 14, 1524. https://doi.org/10.3390/rs14061524 https://www.mdpi.com/journal/remotesensing
