Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning

Sarah Kentsch; Savvas Karatsiolis; Andreas Kamilaris; Luca Tomhave; Maximo Larry Lopez Caceres

doi:10.1007/978-3-030-61969-5_18

Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning

Sarah Kentsch, Savvas Karatsiolis^*, Andreas Kamilaris, Luca Tomhave, Maximo Larry Lopez Caceres

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

19 Downloads (Pure)

Abstract

Natural forests are complex ecosystems whose tree species distribution and their ecosystem functions are still not well understood. Sustainable management of these forests is of high importance because of their significant role in climate regulation, biodiversity, soil erosion and disaster prevention among many other ecosystem services they provide. In Japan particularly, natural forests are mainly located in steep mountains, hence the use of aerial imagery in combination with computer vision are important modern tools that can be applied to forest research. Thus, this study constitutes a preliminary research in this field, aiming at classifying tree species in Japanese mixed forests using UAV images and deep learning in two different mixed forest types: a black pine (Pinus thunbergii)-black locust (Robinia pseudoacacia) and a larch (Larix kaempferi)-oak (Quercus mongolica) mixed forest. Our results indicate that it is possible to identify black locust trees with 62.6% True Positives (TP) and 98.1% True Negatives (TN), while lower precision was reached for larch trees (37.4% TP and 97.7% TN).

Original language	English
Title of host publication	Advances and New Trends in Environmental Informatics
Subtitle of host publication	Digital Twins for Sustainability
Editors	Andreas Kamilaris, Volker Wohlgemuth, Kostas Karatzas, Ioannis N. Athanasiadis
Place of Publication	Cham
Publisher	Springer
Pages	255–270
ISBN (Electronic)	978-3-030-61969-5
ISBN (Print)	978-3-030-61968-8, 978-3-030-61971-8
DOIs	https://doi.org/10.1007/978-3-030-61969-5_18
Publication status	Published - 2021
Event	34th International Conference on Environmental Information and Communication Technologies, EnviroInfo 2020 - Nicosia, Cyprus Duration: 23 Sept 2020 → 24 Sept 2020

Publication series

Name	Progress in IS
Publisher	Springer
ISSN (Print)	2196-8705
ISSN (Electronic)	2196-8713

Conference

Conference	34th International Conference on Environmental Information and Communication Technologies, EnviroInfo 2020
Country/Territory	Cyprus
City	Nicosia
Period	23/09/20 → 24/09/20

Keywords

2024 OA procedure

Access to Document

10.1007/978-3-030-61969-5_18

ArticleFinal published version, 548 KBLicence: Taverne

1 Preprint

Identification of Tree Species in Japanese Forests based on Aerial Photography and Deep Learning
Kentsch, S., Karatsiolis, S., Kamilaris, A., Tomhave, L. & Lopez Caceres, M. L., 17 Jul 2020, ArXiv.org, 15 p.
Research output: Working paper › Preprint › Academic

Open Access
File
5 Downloads (Pure)

Cite this

Kentsch, S., Karatsiolis, S., Kamilaris, A., Tomhave, L., & Lopez Caceres, M. L. (2021). Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning. In A. Kamilaris, V. Wohlgemuth, K. Karatzas, & I. N. Athanasiadis (Eds.), Advances and New Trends in Environmental Informatics: Digital Twins for Sustainability (pp. 255–270). (Progress in IS). Springer. https://doi.org/10.1007/978-3-030-61969-5_18

Kentsch, Sarah ; Karatsiolis, Savvas ; Kamilaris, Andreas et al. / Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning. Advances and New Trends in Environmental Informatics: Digital Twins for Sustainability. editor / Andreas Kamilaris ; Volker Wohlgemuth ; Kostas Karatzas ; Ioannis N. Athanasiadis. Cham : Springer, 2021. pp. 255–270 (Progress in IS).

@inproceedings{b910e1b26c83423493266f29a251e1fa,

title = "Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning",

abstract = "Natural forests are complex ecosystems whose tree species distribution and their ecosystem functions are still not well understood. Sustainable management of these forests is of high importance because of their significant role in climate regulation, biodiversity, soil erosion and disaster prevention among many other ecosystem services they provide. In Japan particularly, natural forests are mainly located in steep mountains, hence the use of aerial imagery in combination with computer vision are important modern tools that can be applied to forest research. Thus, this study constitutes a preliminary research in this field, aiming at classifying tree species in Japanese mixed forests using UAV images and deep learning in two different mixed forest types: a black pine (Pinus thunbergii)-black locust (Robinia pseudoacacia) and a larch (Larix kaempferi)-oak (Quercus mongolica) mixed forest. Our results indicate that it is possible to identify black locust trees with 62.6% True Positives (TP) and 98.1% True Negatives (TN), while lower precision was reached for larch trees (37.4% TP and 97.7% TN).",

keywords = "2024 OA procedure",

author = "Sarah Kentsch and Savvas Karatsiolis and Andreas Kamilaris and Luca Tomhave and {Lopez Caceres}, {Maximo Larry}",

year = "2021",

doi = "10.1007/978-3-030-61969-5_18",

language = "English",

isbn = "978-3-030-61968-8",

series = "Progress in IS",

publisher = "Springer",

pages = "255–270",

editor = "Andreas Kamilaris and Volker Wohlgemuth and Kostas Karatzas and Athanasiadis, {Ioannis N.}",

booktitle = "Advances and New Trends in Environmental Informatics",

address = "Germany",

note = "34th International Conference on Environmental Information and Communication Technologies, EnviroInfo 2020 ; Conference date: 23-09-2020 Through 24-09-2020",

}

Kentsch, S, Karatsiolis, S, Kamilaris, A, Tomhave, L & Lopez Caceres, ML 2021, Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning. in A Kamilaris, V Wohlgemuth, K Karatzas & IN Athanasiadis (eds), Advances and New Trends in Environmental Informatics: Digital Twins for Sustainability. Progress in IS, Springer, Cham, pp. 255–270, 34th International Conference on Environmental Information and Communication Technologies, EnviroInfo 2020, Nicosia, Cyprus, 23/09/20. https://doi.org/10.1007/978-3-030-61969-5_18

Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning. / Kentsch, Sarah; Karatsiolis, Savvas; Kamilaris, Andreas et al.
Advances and New Trends in Environmental Informatics: Digital Twins for Sustainability. ed. / Andreas Kamilaris; Volker Wohlgemuth; Kostas Karatzas; Ioannis N. Athanasiadis. Cham: Springer, 2021. p. 255–270 (Progress in IS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning

AU - Kentsch, Sarah

AU - Karatsiolis, Savvas

AU - Kamilaris, Andreas

AU - Tomhave, Luca

AU - Lopez Caceres, Maximo Larry

PY - 2021

Y1 - 2021

N2 - Natural forests are complex ecosystems whose tree species distribution and their ecosystem functions are still not well understood. Sustainable management of these forests is of high importance because of their significant role in climate regulation, biodiversity, soil erosion and disaster prevention among many other ecosystem services they provide. In Japan particularly, natural forests are mainly located in steep mountains, hence the use of aerial imagery in combination with computer vision are important modern tools that can be applied to forest research. Thus, this study constitutes a preliminary research in this field, aiming at classifying tree species in Japanese mixed forests using UAV images and deep learning in two different mixed forest types: a black pine (Pinus thunbergii)-black locust (Robinia pseudoacacia) and a larch (Larix kaempferi)-oak (Quercus mongolica) mixed forest. Our results indicate that it is possible to identify black locust trees with 62.6% True Positives (TP) and 98.1% True Negatives (TN), while lower precision was reached for larch trees (37.4% TP and 97.7% TN).

AB - Natural forests are complex ecosystems whose tree species distribution and their ecosystem functions are still not well understood. Sustainable management of these forests is of high importance because of their significant role in climate regulation, biodiversity, soil erosion and disaster prevention among many other ecosystem services they provide. In Japan particularly, natural forests are mainly located in steep mountains, hence the use of aerial imagery in combination with computer vision are important modern tools that can be applied to forest research. Thus, this study constitutes a preliminary research in this field, aiming at classifying tree species in Japanese mixed forests using UAV images and deep learning in two different mixed forest types: a black pine (Pinus thunbergii)-black locust (Robinia pseudoacacia) and a larch (Larix kaempferi)-oak (Quercus mongolica) mixed forest. Our results indicate that it is possible to identify black locust trees with 62.6% True Positives (TP) and 98.1% True Negatives (TN), while lower precision was reached for larch trees (37.4% TP and 97.7% TN).

KW - 2024 OA procedure

U2 - 10.1007/978-3-030-61969-5_18

DO - 10.1007/978-3-030-61969-5_18

M3 - Conference contribution

SN - 978-3-030-61968-8

SN - 978-3-030-61971-8

T3 - Progress in IS

SP - 255

EP - 270

BT - Advances and New Trends in Environmental Informatics

A2 - Kamilaris, Andreas

A2 - Wohlgemuth, Volker

A2 - Karatzas, Kostas

A2 - Athanasiadis, Ioannis N.

PB - Springer

CY - Cham

T2 - 34th International Conference on Environmental Information and Communication Technologies, EnviroInfo 2020

Y2 - 23 September 2020 through 24 September 2020

ER -

Kentsch S, Karatsiolis S, Kamilaris A, Tomhave L, Lopez Caceres ML. Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning. In Kamilaris A, Wohlgemuth V, Karatzas K, Athanasiadis IN, editors, Advances and New Trends in Environmental Informatics: Digital Twins for Sustainability. Cham: Springer. 2021. p. 255–270. (Progress in IS). Epub 2020 Dec 17. doi: 10.1007/978-3-030-61969-5_18

Identification of Tree Species in Japanese Forests Based on Aerial Photography and Deep Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Research output

Identification of Tree Species in Japanese Forests based on Aerial Photography and Deep Learning

Cite this