Classification of tree species from 3D point clouds using convolutional neural networks
In forest management, knowledge about a forest's distribution of tree species is key. Being able to automate tree species classification for large forest areas is of great interest, since it is tedious and costly labour doing it manually. In this project, the aim was to investigate the efficien...
Main Author: | |
---|---|
Format: | Bachelor Thesis |
Language: | English |
Published: |
Umeå universitet, Institutionen för fysik
2020
|
Subjects: | |
Online Access: | http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-174662 |
id |
ftumeauniv:oai:DiVA.org:umu-174662 |
---|---|
record_format |
openpolar |
spelling |
ftumeauniv:oai:DiVA.org:umu-174662 2023-10-09T21:54:37+02:00 Classification of tree species from 3D point clouds using convolutional neural networks Wiklander, Marcus 2020 application/pdf http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-174662 eng eng Umeå universitet, Institutionen för fysik http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-174662 info:eu-repo/semantics/openAccess Neural networks Convolutional neural networks tree species classification Forest Science Skogsvetenskap Computer Vision and Robotics (Autonomous Systems) Datorseende och robotik (autonoma system) Student thesis info:eu-repo/semantics/bachelorThesis text 2020 ftumeauniv 2023-09-22T13:53:03Z In forest management, knowledge about a forest's distribution of tree species is key. Being able to automate tree species classification for large forest areas is of great interest, since it is tedious and costly labour doing it manually. In this project, the aim was to investigate the efficiency of classifying individual tree species (pine, spruce and deciduous forest) from 3D point clouds acquired by airborne laser scanning (ALS), using convolutional neural networks. Raw data consisted of 3D point clouds and photographic images of forests in northern Sweden, collected from a helicopter flying at low altitudes. The point cloud of each individual tree was connected to its representation in the photos, which allowed for manual labeling of training data to be used for training of convolutional neural networks. The training data consisted of labels and 2D projections created from the point clouds, represented as images. Two different convolutional neural networks were trained and tested; an adaptation of the LeNet architecture and the ResNet architecture. Both networks reached an accuracy close to 98 %, the LeNet adaptation having a slightly lower loss score for both validation and test data compared to that of ResNet. Confusion matrices for both networks showed similar F1 scores for all tree species, between 97 % and 98 %. The accuracies computed for both networks were found higher than those achieved in similar studies using ALS data to classify individual tree species. However, the results in this project were never tested against a true population sample to confirm the accuracy. To conclude, the use of convolutional neural networks is indeed an efficient method for classification of tree species, but further studies on unbiased data is needed to validate these results. Bachelor Thesis Northern Sweden Umeå University: Publications (DiVA) |
institution |
Open Polar |
collection |
Umeå University: Publications (DiVA) |
op_collection_id |
ftumeauniv |
language |
English |
topic |
Neural networks Convolutional neural networks tree species classification Forest Science Skogsvetenskap Computer Vision and Robotics (Autonomous Systems) Datorseende och robotik (autonoma system) |
spellingShingle |
Neural networks Convolutional neural networks tree species classification Forest Science Skogsvetenskap Computer Vision and Robotics (Autonomous Systems) Datorseende och robotik (autonoma system) Wiklander, Marcus Classification of tree species from 3D point clouds using convolutional neural networks |
topic_facet |
Neural networks Convolutional neural networks tree species classification Forest Science Skogsvetenskap Computer Vision and Robotics (Autonomous Systems) Datorseende och robotik (autonoma system) |
description |
In forest management, knowledge about a forest's distribution of tree species is key. Being able to automate tree species classification for large forest areas is of great interest, since it is tedious and costly labour doing it manually. In this project, the aim was to investigate the efficiency of classifying individual tree species (pine, spruce and deciduous forest) from 3D point clouds acquired by airborne laser scanning (ALS), using convolutional neural networks. Raw data consisted of 3D point clouds and photographic images of forests in northern Sweden, collected from a helicopter flying at low altitudes. The point cloud of each individual tree was connected to its representation in the photos, which allowed for manual labeling of training data to be used for training of convolutional neural networks. The training data consisted of labels and 2D projections created from the point clouds, represented as images. Two different convolutional neural networks were trained and tested; an adaptation of the LeNet architecture and the ResNet architecture. Both networks reached an accuracy close to 98 %, the LeNet adaptation having a slightly lower loss score for both validation and test data compared to that of ResNet. Confusion matrices for both networks showed similar F1 scores for all tree species, between 97 % and 98 %. The accuracies computed for both networks were found higher than those achieved in similar studies using ALS data to classify individual tree species. However, the results in this project were never tested against a true population sample to confirm the accuracy. To conclude, the use of convolutional neural networks is indeed an efficient method for classification of tree species, but further studies on unbiased data is needed to validate these results. |
format |
Bachelor Thesis |
author |
Wiklander, Marcus |
author_facet |
Wiklander, Marcus |
author_sort |
Wiklander, Marcus |
title |
Classification of tree species from 3D point clouds using convolutional neural networks |
title_short |
Classification of tree species from 3D point clouds using convolutional neural networks |
title_full |
Classification of tree species from 3D point clouds using convolutional neural networks |
title_fullStr |
Classification of tree species from 3D point clouds using convolutional neural networks |
title_full_unstemmed |
Classification of tree species from 3D point clouds using convolutional neural networks |
title_sort |
classification of tree species from 3d point clouds using convolutional neural networks |
publisher |
Umeå universitet, Institutionen för fysik |
publishDate |
2020 |
url |
http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-174662 |
genre |
Northern Sweden |
genre_facet |
Northern Sweden |
op_relation |
http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-174662 |
op_rights |
info:eu-repo/semantics/openAccess |
_version_ |
1779318261349875712 |