Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression
Most of the allometric models used to estimate tree aboveground biomass rely on tree diameter at breast height (DBH). However, it is difficult to measure DBH from airborne remote sensors, and is common to draw upon traditional least squares linear regression models to relate DBH with dendrometric va...
| Main Authors: | , , , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
MDPI
2021
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10835/12779 |
| _version_ | 1789406313540747264 |
|---|---|
| author | Aguilar Torres, Fernando José Nemmaoui, Abderrahim Aguilar Torres, Manuel Ángel Peñalver, Alberto |
| author_facet | Aguilar Torres, Fernando José Nemmaoui, Abderrahim Aguilar Torres, Manuel Ángel Peñalver, Alberto |
| author_sort | Aguilar Torres, Fernando José |
| collection | DSpace |
| description | Most of the allometric models used to estimate tree aboveground biomass rely on tree diameter at breast height (DBH). However, it is difficult to measure DBH from airborne remote sensors, and is common to draw upon traditional least squares linear regression models to relate DBH with dendrometric variables measured from airborne sensors, such as tree height (H) and crown diameter (CD). This study explores the usefulness of ensemble-type supervised machine learning regression algorithms, such as random forest regression (RFR), categorical boosting (CatBoost), gradient boosting (GBoost), or AdaBoost regression (AdaBoost), as an alternative to linear regression (LR) for modelling the allometric relationships DBH = Φ(H) and DBH = Ψ(H, CD). The original dataset was made up of 2272 teak trees (Tectona grandis Linn. F.) belonging to three different plantations located in Ecuador. All teak trees were digitally reconstructed from terrestrial laser scanning point clouds. The results showed that allometric models involving both H and CD to estimate DBH performed better than those based solely on H. Furthermore, boosting machine learning regression algorithms (CatBoost and GBoost) outperformed RFR (bagging) and LR (traditional linear regression) models, both in terms of goodness-of-fit (R2) and stability (variations in training and testing samples). |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-12779 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2021 |
| publisher | MDPI |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-127792023-10-10T11:07:36Z Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression Aguilar Torres, Fernando José Nemmaoui, Abderrahim Aguilar Torres, Manuel Ángel Peñalver, Alberto terrestrial laser scanning allometricmodels machine learning regression teak plantations forest inventory Most of the allometric models used to estimate tree aboveground biomass rely on tree diameter at breast height (DBH). However, it is difficult to measure DBH from airborne remote sensors, and is common to draw upon traditional least squares linear regression models to relate DBH with dendrometric variables measured from airborne sensors, such as tree height (H) and crown diameter (CD). This study explores the usefulness of ensemble-type supervised machine learning regression algorithms, such as random forest regression (RFR), categorical boosting (CatBoost), gradient boosting (GBoost), or AdaBoost regression (AdaBoost), as an alternative to linear regression (LR) for modelling the allometric relationships DBH = Φ(H) and DBH = Ψ(H, CD). The original dataset was made up of 2272 teak trees (Tectona grandis Linn. F.) belonging to three different plantations located in Ecuador. All teak trees were digitally reconstructed from terrestrial laser scanning point clouds. The results showed that allometric models involving both H and CD to estimate DBH performed better than those based solely on H. Furthermore, boosting machine learning regression algorithms (CatBoost and GBoost) outperformed RFR (bagging) and LR (traditional linear regression) models, both in terms of goodness-of-fit (R2) and stability (variations in training and testing samples). 2021-11-15T11:29:54Z 2021-11-15T11:29:54Z 2021-10-29 info:eu-repo/semantics/article 2076-3417 http://hdl.handle.net/10835/12779 10.3390/app112110139 en Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess MDPI |
| spellingShingle | terrestrial laser scanning allometricmodels machine learning regression teak plantations forest inventory Aguilar Torres, Fernando José Nemmaoui, Abderrahim Aguilar Torres, Manuel Ángel Peñalver, Alberto Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression |
| title | Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression |
| title_full | Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression |
| title_fullStr | Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression |
| title_full_unstemmed | Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression |
| title_short | Building Tree Allometry Relationships Based on TLS Point Clouds and Machine Learning Regression |
| title_sort | building tree allometry relationships based on tls point clouds and machine learning regression |
| topic | terrestrial laser scanning allometricmodels machine learning regression teak plantations forest inventory |
| url | http://hdl.handle.net/10835/12779 |
| work_keys_str_mv | AT aguilartorresfernandojose buildingtreeallometryrelationshipsbasedontlspointcloudsandmachinelearningregression AT nemmaouiabderrahim buildingtreeallometryrelationshipsbasedontlspointcloudsandmachinelearningregression AT aguilartorresmanuelangel buildingtreeallometryrelationshipsbasedontlspointcloudsandmachinelearningregression AT penalveralberto buildingtreeallometryrelationshipsbasedontlspointcloudsandmachinelearningregression |