Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging

Olli Nevalainen (Corresponding Author), Eija Honkavaara, Sakari Tuominen, Niko Viljanen, Teemu Hakala, Xiaowei Yu, Juha Hyyppä, Heikki Saari, Ilkka Pölönen, Nilton N. Imai, Antonio M.G. Tommaselli

Research output: Contribution to journalArticleScientificpeer-review

75 Citations (Scopus)

Abstract

Small unmanned aerial vehicle (UAV) based remote sensing is a rapidly evolving technology. Novel sensors and methods are entering the market, offering completely new possibilities to carry out remote sensing tasks. Three-dimensional (3D) hyperspectral remote sensing is a novel and powerful technology that has recently become available to small UAVs. This study investigated the performance of UAV-based photogrammetry and hyperspectral imaging in individual tree detection and tree species classification in boreal forests. Eleven test sites with 4151 reference trees representing various tree species and developmental stages were collected in June 2014 using a UAV remote sensing system equipped with a frame format hyperspectral camera and an RGB camera in highly variable weather conditions. Dense point clouds were measured photogrammetrically by automatic image matching using high resolution RGB images with a 5 cm point interval. Spectral features were obtained from the hyperspectral image blocks, the large radiometric variation of which was compensated for by using a novel approach based on radiometric block adjustment with the support of in-flight irradiance observations. Spectral and 3D point cloud features were used in the classification experiment with various classifiers. The best results were obtained with Random Forest and Multilayer Perceptron (MLP) which both gave 95% overall accuracies and an F-score of 0.93. Accuracy of individual tree identification from the photogrammetric point clouds varied between 40% and 95%, depending on the characteristics of the area. Challenges in reference measurements might also have reduced these numbers. Results were promising, indicating that hyperspectral 3D remote sensing was operational from a UAV platform even in very difficult conditions. These novel methods are expected to provide a powerful tool for automating various environmental close-range remote sensing tasks in the very near future.
Original languageEnglish
Article number185
JournalRemote Sensing
Volume9
Issue number3
DOIs
Publication statusPublished - 1 Mar 2017
MoE publication typeA1 Journal article-refereed

Fingerprint

remote sensing
image resolution
photogrammetry
developmental stage
boreal forest
irradiance
detection
vehicle
flight
sensor
market
experiment
method

Keywords

  • classification
  • forest
  • hyperspectral
  • photogrammetry
  • point cloud
  • radiometry
  • UAV

Cite this

Nevalainen, O., Honkavaara, E., Tuominen, S., Viljanen, N., Hakala, T., Yu, X., ... Tommaselli, A. M. G. (2017). Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging. Remote Sensing, 9(3), [185]. https://doi.org/10.3390/rs9030185
Nevalainen, Olli ; Honkavaara, Eija ; Tuominen, Sakari ; Viljanen, Niko ; Hakala, Teemu ; Yu, Xiaowei ; Hyyppä, Juha ; Saari, Heikki ; Pölönen, Ilkka ; Imai, Nilton N. ; Tommaselli, Antonio M.G. / Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging. In: Remote Sensing. 2017 ; Vol. 9, No. 3.
@article{82389dc4801c4cc89942bf7bc3241ae7,
title = "Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging",
abstract = "Small unmanned aerial vehicle (UAV) based remote sensing is a rapidly evolving technology. Novel sensors and methods are entering the market, offering completely new possibilities to carry out remote sensing tasks. Three-dimensional (3D) hyperspectral remote sensing is a novel and powerful technology that has recently become available to small UAVs. This study investigated the performance of UAV-based photogrammetry and hyperspectral imaging in individual tree detection and tree species classification in boreal forests. Eleven test sites with 4151 reference trees representing various tree species and developmental stages were collected in June 2014 using a UAV remote sensing system equipped with a frame format hyperspectral camera and an RGB camera in highly variable weather conditions. Dense point clouds were measured photogrammetrically by automatic image matching using high resolution RGB images with a 5 cm point interval. Spectral features were obtained from the hyperspectral image blocks, the large radiometric variation of which was compensated for by using a novel approach based on radiometric block adjustment with the support of in-flight irradiance observations. Spectral and 3D point cloud features were used in the classification experiment with various classifiers. The best results were obtained with Random Forest and Multilayer Perceptron (MLP) which both gave 95{\%} overall accuracies and an F-score of 0.93. Accuracy of individual tree identification from the photogrammetric point clouds varied between 40{\%} and 95{\%}, depending on the characteristics of the area. Challenges in reference measurements might also have reduced these numbers. Results were promising, indicating that hyperspectral 3D remote sensing was operational from a UAV platform even in very difficult conditions. These novel methods are expected to provide a powerful tool for automating various environmental close-range remote sensing tasks in the very near future.",
keywords = "classification, forest, hyperspectral, photogrammetry, point cloud, radiometry, UAV",
author = "Olli Nevalainen and Eija Honkavaara and Sakari Tuominen and Niko Viljanen and Teemu Hakala and Xiaowei Yu and Juha Hyypp{\"a} and Heikki Saari and Ilkka P{\"o}l{\"o}nen and Imai, {Nilton N.} and Tommaselli, {Antonio M.G.}",
year = "2017",
month = "3",
day = "1",
doi = "10.3390/rs9030185",
language = "English",
volume = "9",
journal = "Remote Sensing",
issn = "2072-4292",
publisher = "MDPI",
number = "3",

}

Nevalainen, O, Honkavaara, E, Tuominen, S, Viljanen, N, Hakala, T, Yu, X, Hyyppä, J, Saari, H, Pölönen, I, Imai, NN & Tommaselli, AMG 2017, 'Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging', Remote Sensing, vol. 9, no. 3, 185. https://doi.org/10.3390/rs9030185

Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging. / Nevalainen, Olli (Corresponding Author); Honkavaara, Eija; Tuominen, Sakari; Viljanen, Niko; Hakala, Teemu; Yu, Xiaowei; Hyyppä, Juha; Saari, Heikki; Pölönen, Ilkka; Imai, Nilton N.; Tommaselli, Antonio M.G.

In: Remote Sensing, Vol. 9, No. 3, 185, 01.03.2017.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging

AU - Nevalainen, Olli

AU - Honkavaara, Eija

AU - Tuominen, Sakari

AU - Viljanen, Niko

AU - Hakala, Teemu

AU - Yu, Xiaowei

AU - Hyyppä, Juha

AU - Saari, Heikki

AU - Pölönen, Ilkka

AU - Imai, Nilton N.

AU - Tommaselli, Antonio M.G.

PY - 2017/3/1

Y1 - 2017/3/1

N2 - Small unmanned aerial vehicle (UAV) based remote sensing is a rapidly evolving technology. Novel sensors and methods are entering the market, offering completely new possibilities to carry out remote sensing tasks. Three-dimensional (3D) hyperspectral remote sensing is a novel and powerful technology that has recently become available to small UAVs. This study investigated the performance of UAV-based photogrammetry and hyperspectral imaging in individual tree detection and tree species classification in boreal forests. Eleven test sites with 4151 reference trees representing various tree species and developmental stages were collected in June 2014 using a UAV remote sensing system equipped with a frame format hyperspectral camera and an RGB camera in highly variable weather conditions. Dense point clouds were measured photogrammetrically by automatic image matching using high resolution RGB images with a 5 cm point interval. Spectral features were obtained from the hyperspectral image blocks, the large radiometric variation of which was compensated for by using a novel approach based on radiometric block adjustment with the support of in-flight irradiance observations. Spectral and 3D point cloud features were used in the classification experiment with various classifiers. The best results were obtained with Random Forest and Multilayer Perceptron (MLP) which both gave 95% overall accuracies and an F-score of 0.93. Accuracy of individual tree identification from the photogrammetric point clouds varied between 40% and 95%, depending on the characteristics of the area. Challenges in reference measurements might also have reduced these numbers. Results were promising, indicating that hyperspectral 3D remote sensing was operational from a UAV platform even in very difficult conditions. These novel methods are expected to provide a powerful tool for automating various environmental close-range remote sensing tasks in the very near future.

AB - Small unmanned aerial vehicle (UAV) based remote sensing is a rapidly evolving technology. Novel sensors and methods are entering the market, offering completely new possibilities to carry out remote sensing tasks. Three-dimensional (3D) hyperspectral remote sensing is a novel and powerful technology that has recently become available to small UAVs. This study investigated the performance of UAV-based photogrammetry and hyperspectral imaging in individual tree detection and tree species classification in boreal forests. Eleven test sites with 4151 reference trees representing various tree species and developmental stages were collected in June 2014 using a UAV remote sensing system equipped with a frame format hyperspectral camera and an RGB camera in highly variable weather conditions. Dense point clouds were measured photogrammetrically by automatic image matching using high resolution RGB images with a 5 cm point interval. Spectral features were obtained from the hyperspectral image blocks, the large radiometric variation of which was compensated for by using a novel approach based on radiometric block adjustment with the support of in-flight irradiance observations. Spectral and 3D point cloud features were used in the classification experiment with various classifiers. The best results were obtained with Random Forest and Multilayer Perceptron (MLP) which both gave 95% overall accuracies and an F-score of 0.93. Accuracy of individual tree identification from the photogrammetric point clouds varied between 40% and 95%, depending on the characteristics of the area. Challenges in reference measurements might also have reduced these numbers. Results were promising, indicating that hyperspectral 3D remote sensing was operational from a UAV platform even in very difficult conditions. These novel methods are expected to provide a powerful tool for automating various environmental close-range remote sensing tasks in the very near future.

KW - classification

KW - forest

KW - hyperspectral

KW - photogrammetry

KW - point cloud

KW - radiometry

KW - UAV

UR - http://www.scopus.com/inward/record.url?scp=85019363955&partnerID=8YFLogxK

U2 - 10.3390/rs9030185

DO - 10.3390/rs9030185

M3 - Article

VL - 9

JO - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 3

M1 - 185

ER -