Cargando…

Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning

Automated species classification from 3D point clouds is still a challenge. It is, however, an important task for laser scanning-based forest inventory, ecosystem models, and to support forest management. Here, we tested the performance of an image classification approach based on convolutional neur...

Descripción completa

Detalles Bibliográficos
Autores principales: Seidel, Dominik, Annighöfer, Peter, Thielman, Anton, Seifert, Quentin Edward, Thauer, Jan-Henrik, Glatthorn, Jonas, Ehbrecht, Martin, Kneib, Thomas, Ammer, Christian
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7902704/
https://www.ncbi.nlm.nih.gov/pubmed/33643364
http://dx.doi.org/10.3389/fpls.2021.635440
_version_ 1783654584533647360
author Seidel, Dominik
Annighöfer, Peter
Thielman, Anton
Seifert, Quentin Edward
Thauer, Jan-Henrik
Glatthorn, Jonas
Ehbrecht, Martin
Kneib, Thomas
Ammer, Christian
author_facet Seidel, Dominik
Annighöfer, Peter
Thielman, Anton
Seifert, Quentin Edward
Thauer, Jan-Henrik
Glatthorn, Jonas
Ehbrecht, Martin
Kneib, Thomas
Ammer, Christian
author_sort Seidel, Dominik
collection PubMed
description Automated species classification from 3D point clouds is still a challenge. It is, however, an important task for laser scanning-based forest inventory, ecosystem models, and to support forest management. Here, we tested the performance of an image classification approach based on convolutional neural networks (CNNs) with the aim to classify 3D point clouds of seven tree species based on 2D representation in a computationally efficient way. We were particularly interested in how the approach would perform with artificially increased training data size based on image augmentation techniques. Our approach yielded a high classification accuracy (86%) and the confusion matrix revealed that despite rather small sample sizes of the training data for some tree species, classification accuracy was high. We could partly relate this to the successful application of the image augmentation technique, improving our result by 6% in total and 13, 14, and 24% for ash, oak and pine, respectively. The introduced approach is hence not only applicable to small-sized datasets, it is also computationally effective since it relies on 2D instead of 3D data to be processed in the CNN. Our approach was faster and more accurate when compared to the point cloud-based “PointNet” approach.
format Online
Article
Text
id pubmed-7902704
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-79027042021-02-25 Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning Seidel, Dominik Annighöfer, Peter Thielman, Anton Seifert, Quentin Edward Thauer, Jan-Henrik Glatthorn, Jonas Ehbrecht, Martin Kneib, Thomas Ammer, Christian Front Plant Sci Plant Science Automated species classification from 3D point clouds is still a challenge. It is, however, an important task for laser scanning-based forest inventory, ecosystem models, and to support forest management. Here, we tested the performance of an image classification approach based on convolutional neural networks (CNNs) with the aim to classify 3D point clouds of seven tree species based on 2D representation in a computationally efficient way. We were particularly interested in how the approach would perform with artificially increased training data size based on image augmentation techniques. Our approach yielded a high classification accuracy (86%) and the confusion matrix revealed that despite rather small sample sizes of the training data for some tree species, classification accuracy was high. We could partly relate this to the successful application of the image augmentation technique, improving our result by 6% in total and 13, 14, and 24% for ash, oak and pine, respectively. The introduced approach is hence not only applicable to small-sized datasets, it is also computationally effective since it relies on 2D instead of 3D data to be processed in the CNN. Our approach was faster and more accurate when compared to the point cloud-based “PointNet” approach. Frontiers Media S.A. 2021-02-10 /pmc/articles/PMC7902704/ /pubmed/33643364 http://dx.doi.org/10.3389/fpls.2021.635440 Text en Copyright © 2021 Seidel, Annighöfer, Thielman, Seifert, Thauer, Glatthorn, Ehbrecht, Kneib and Ammer. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Plant Science
Seidel, Dominik
Annighöfer, Peter
Thielman, Anton
Seifert, Quentin Edward
Thauer, Jan-Henrik
Glatthorn, Jonas
Ehbrecht, Martin
Kneib, Thomas
Ammer, Christian
Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning
title Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning
title_full Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning
title_fullStr Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning
title_full_unstemmed Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning
title_short Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning
title_sort predicting tree species from 3d laser scanning point clouds using deep learning
topic Plant Science
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7902704/
https://www.ncbi.nlm.nih.gov/pubmed/33643364
http://dx.doi.org/10.3389/fpls.2021.635440
work_keys_str_mv AT seideldominik predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT annighoferpeter predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT thielmananton predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT seifertquentinedward predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT thauerjanhenrik predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT glatthornjonas predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT ehbrechtmartin predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT kneibthomas predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning
AT ammerchristian predictingtreespeciesfrom3dlaserscanningpointcloudsusingdeeplearning