Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms
Canopy closure is a critical indicator reflecting forest structure, biodiversity, and ecological balance. This study proposes an estimation method integrating U-Net segmentation with machine learning, significantly improving accuracy through multi-source remote sensing data and feature selection. Co...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/11/1828 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849722336856506368 |
|---|---|
| author | Yuefei Zhou Jinghan Wang Zengjing Song Miaohang Zhou Mengnan Lv Xujun Han |
| author_facet | Yuefei Zhou Jinghan Wang Zengjing Song Miaohang Zhou Mengnan Lv Xujun Han |
| author_sort | Yuefei Zhou |
| collection | DOAJ |
| description | Canopy closure is a critical indicator reflecting forest structure, biodiversity, and ecological balance. This study proposes an estimation method integrating U-Net segmentation with machine learning, significantly improving accuracy through multi-source remote sensing data and feature selection. Covering eight U.S. continental states, the study utilized 13,000 stratified samples equally split for model training and validation. Four states were used to train models based on XGBoost, random forest (RF), LightGBM, and support vector machine (SVM), while the remaining four states served for validation. The results indicate that (1) U-Net effectively extracted tree crowns from aerial imagery to construct the sample dataset; (2) among the tested algorithms, XGBoost achieved the highest accuracy of 0.88 when incorporating Sentinel-1, Sentinel-2, vegetation indices, and land cover features, outperforming models using only Sentinel-2 data by 25.7%; and (3) XGBoost-estimated tree canopy cover (Model TCC) showed finer spatial details than the National Land Cover Database Tree Canopy Cover (NLCD TCC), with R<sup>2</sup> against the true tree canopy closure from U-Net (True TCC) up to 49.1% higher. This approach offers a cost-effective solution for regional-scale canopy monitoring. |
| format | Article |
| id | doaj-art-ad58abbd30864032ae50dd4a970dcbe2 |
| institution | DOAJ |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-ad58abbd30864032ae50dd4a970dcbe22025-08-20T03:11:22ZengMDPI AGRemote Sensing2072-42922025-05-011711182810.3390/rs17111828Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning AlgorithmsYuefei Zhou0Jinghan Wang1Zengjing Song2Miaohang Zhou3Mengnan Lv4Xujun Han5Chongqing Engineering Research Center for Remote Sensing Big Data Application, Chongqing Jinfo Mountain National Field Scientific Observation and Research Station for Karst Ecosystem, School of Geographical Sciences, Southwest University, Chongqing 400715, ChinaChongqing Engineering Research Center for Remote Sensing Big Data Application, Chongqing Jinfo Mountain National Field Scientific Observation and Research Station for Karst Ecosystem, School of Geographical Sciences, Southwest University, Chongqing 400715, ChinaChongqing Engineering Research Center for Remote Sensing Big Data Application, Chongqing Jinfo Mountain National Field Scientific Observation and Research Station for Karst Ecosystem, School of Geographical Sciences, Southwest University, Chongqing 400715, ChinaChongqing Engineering Research Center for Remote Sensing Big Data Application, Chongqing Jinfo Mountain National Field Scientific Observation and Research Station for Karst Ecosystem, School of Geographical Sciences, Southwest University, Chongqing 400715, ChinaChongqing Engineering Research Center for Remote Sensing Big Data Application, Chongqing Jinfo Mountain National Field Scientific Observation and Research Station for Karst Ecosystem, School of Geographical Sciences, Southwest University, Chongqing 400715, ChinaChongqing Engineering Research Center for Remote Sensing Big Data Application, Chongqing Jinfo Mountain National Field Scientific Observation and Research Station for Karst Ecosystem, School of Geographical Sciences, Southwest University, Chongqing 400715, ChinaCanopy closure is a critical indicator reflecting forest structure, biodiversity, and ecological balance. This study proposes an estimation method integrating U-Net segmentation with machine learning, significantly improving accuracy through multi-source remote sensing data and feature selection. Covering eight U.S. continental states, the study utilized 13,000 stratified samples equally split for model training and validation. Four states were used to train models based on XGBoost, random forest (RF), LightGBM, and support vector machine (SVM), while the remaining four states served for validation. The results indicate that (1) U-Net effectively extracted tree crowns from aerial imagery to construct the sample dataset; (2) among the tested algorithms, XGBoost achieved the highest accuracy of 0.88 when incorporating Sentinel-1, Sentinel-2, vegetation indices, and land cover features, outperforming models using only Sentinel-2 data by 25.7%; and (3) XGBoost-estimated tree canopy cover (Model TCC) showed finer spatial details than the National Land Cover Database Tree Canopy Cover (NLCD TCC), with R<sup>2</sup> against the true tree canopy closure from U-Net (True TCC) up to 49.1% higher. This approach offers a cost-effective solution for regional-scale canopy monitoring.https://www.mdpi.com/2072-4292/17/11/1828canopy closuremachine learninghigh-resolution aerial imageryU-Net |
| spellingShingle | Yuefei Zhou Jinghan Wang Zengjing Song Miaohang Zhou Mengnan Lv Xujun Han Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms Remote Sensing canopy closure machine learning high-resolution aerial imagery U-Net |
| title | Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms |
| title_full | Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms |
| title_fullStr | Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms |
| title_full_unstemmed | Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms |
| title_short | Estimation of Tree Canopy Closure Based on U-Net Image Segmentation and Machine Learning Algorithms |
| title_sort | estimation of tree canopy closure based on u net image segmentation and machine learning algorithms |
| topic | canopy closure machine learning high-resolution aerial imagery U-Net |
| url | https://www.mdpi.com/2072-4292/17/11/1828 |
| work_keys_str_mv | AT yuefeizhou estimationoftreecanopyclosurebasedonunetimagesegmentationandmachinelearningalgorithms AT jinghanwang estimationoftreecanopyclosurebasedonunetimagesegmentationandmachinelearningalgorithms AT zengjingsong estimationoftreecanopyclosurebasedonunetimagesegmentationandmachinelearningalgorithms AT miaohangzhou estimationoftreecanopyclosurebasedonunetimagesegmentationandmachinelearningalgorithms AT mengnanlv estimationoftreecanopyclosurebasedonunetimagesegmentationandmachinelearningalgorithms AT xujunhan estimationoftreecanopyclosurebasedonunetimagesegmentationandmachinelearningalgorithms |