Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data

Timely and accurate tree species mapping is crucial for forest resource inventory, supporting management, conservation biology, and ecological restoration. This study utilized Sentinel-1 and Sentinel-2 data to classify five dominant tree species in Chengde and Beijing. To effectively capture the inf...

Full description

Saved in:
Bibliographic Details
Main Authors: Heyi Guo, Sornkitja Boonprong, Shaohua Wang, Zhidong Zhang, Wei Liang, Min Xu, Xinwei Yang, Kaimin Wang, Jingbo Li, Xiaotong Gao, Yujie Yang, Ruichen Hu, Yu Zhang, Chunxiang Cao
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/24/4674
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850059990798172160
author Heyi Guo
Sornkitja Boonprong
Shaohua Wang
Zhidong Zhang
Wei Liang
Min Xu
Xinwei Yang
Kaimin Wang
Jingbo Li
Xiaotong Gao
Yujie Yang
Ruichen Hu
Yu Zhang
Chunxiang Cao
author_facet Heyi Guo
Sornkitja Boonprong
Shaohua Wang
Zhidong Zhang
Wei Liang
Min Xu
Xinwei Yang
Kaimin Wang
Jingbo Li
Xiaotong Gao
Yujie Yang
Ruichen Hu
Yu Zhang
Chunxiang Cao
author_sort Heyi Guo
collection DOAJ
description Timely and accurate tree species mapping is crucial for forest resource inventory, supporting management, conservation biology, and ecological restoration. This study utilized Sentinel-1 and Sentinel-2 data to classify five dominant tree species in Chengde and Beijing. To effectively capture the influence of multi-temporal data, data were acquired in March, June, September, and December 2020, extracting various features, including bands, spectral indices, texture features, and topographic variables. The optimal input variable combination was explored using 1519 field survey samples for training and testing datasets. Classification employed Random Forest, XGBoost, and deep learning models, with performance evaluated through out-of-bag estimation and cross-validation. The XGBoost model achieved the highest accuracy of 81.25% (kappa = 0.74) when using Sentinel-1 and Sentinel-2 bands, indices, texture features, and DEM data. Results demonstrate the effectiveness of using Sentinel data for tree species classification and emphasize the value of machine learning algorithms. This study underscores the potential of combining synthetic aperture radar (SAR) and optical data for large-scale tree species classification, with significant implications for forest monitoring and management.
format Article
id doaj-art-9e854f032f134694be62c62c5e0ed3b3
institution DOAJ
issn 2072-4292
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-9e854f032f134694be62c62c5e0ed3b32025-08-20T02:50:43ZengMDPI AGRemote Sensing2072-42922024-12-011624467410.3390/rs16244674Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source DataHeyi Guo0Sornkitja Boonprong1Shaohua Wang2Zhidong Zhang3Wei Liang4Min Xu5Xinwei Yang6Kaimin Wang7Jingbo Li8Xiaotong Gao9Yujie Yang10Ruichen Hu11Yu Zhang12Chunxiang Cao13Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaFaculty of Social Sciences, Kasetsart University, Bangkok 10900, ThailandKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAcademy of Forestry Inventory and Planning, National Forestry and Grassland Administration, Beijing 100714, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaTimely and accurate tree species mapping is crucial for forest resource inventory, supporting management, conservation biology, and ecological restoration. This study utilized Sentinel-1 and Sentinel-2 data to classify five dominant tree species in Chengde and Beijing. To effectively capture the influence of multi-temporal data, data were acquired in March, June, September, and December 2020, extracting various features, including bands, spectral indices, texture features, and topographic variables. The optimal input variable combination was explored using 1519 field survey samples for training and testing datasets. Classification employed Random Forest, XGBoost, and deep learning models, with performance evaluated through out-of-bag estimation and cross-validation. The XGBoost model achieved the highest accuracy of 81.25% (kappa = 0.74) when using Sentinel-1 and Sentinel-2 bands, indices, texture features, and DEM data. Results demonstrate the effectiveness of using Sentinel data for tree species classification and emphasize the value of machine learning algorithms. This study underscores the potential of combining synthetic aperture radar (SAR) and optical data for large-scale tree species classification, with significant implications for forest monitoring and management.https://www.mdpi.com/2072-4292/16/24/4674tree speciessentinel-1/2random forest, XGBoostdeep learninglarge scale mapping
spellingShingle Heyi Guo
Sornkitja Boonprong
Shaohua Wang
Zhidong Zhang
Wei Liang
Min Xu
Xinwei Yang
Kaimin Wang
Jingbo Li
Xiaotong Gao
Yujie Yang
Ruichen Hu
Yu Zhang
Chunxiang Cao
Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data
Remote Sensing
tree species
sentinel-1/2
random forest, XGBoost
deep learning
large scale mapping
title Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data
title_full Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data
title_fullStr Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data
title_full_unstemmed Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data
title_short Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data
title_sort dominant tree species mapping using machine learning based on multi temporal and multi source data
topic tree species
sentinel-1/2
random forest, XGBoost
deep learning
large scale mapping
url https://www.mdpi.com/2072-4292/16/24/4674
work_keys_str_mv AT heyiguo dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT sornkitjaboonprong dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT shaohuawang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT zhidongzhang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT weiliang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT minxu dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT xinweiyang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT kaiminwang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT jingboli dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT xiaotonggao dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT yujieyang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT ruichenhu dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT yuzhang dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata
AT chunxiangcao dominanttreespeciesmappingusingmachinelearningbasedonmultitemporalandmultisourcedata