A1 Journal article (refereed)
Tree species classification of drone hyperspectral and RGB imagery with deep learning convolutional neural networks (2020)

Nezami, S., Khoramshahi, E., Nevalainen, O., Pölönen, I., & Honkavaara, E. (2020). Tree species classification of drone hyperspectral and RGB imagery with deep learning convolutional neural networks. Remote Sensing, 12(7), Article 1070. https://doi.org/10.3390/rs12071070

JYU authors or editors

Publication details

All authors or editors: Nezami, Somayeh; Khoramshahi, Ehsan; Nevalainen, Olli; Pölönen, Ilkka; Honkavaara, Eija

Journal or series: Remote Sensing

eISSN: 2072-4292

Publication year: 2020

Volume: 12

Issue number: 7

Article number: 1070

Publisher: MDPI AG

Publication country: Switzerland

Publication language: English

DOI: https://doi.org/10.3390/rs12071070

Publication open access: Openly available

Publication channel open access: Open Access channel

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/69060

Publication is parallel published: https://tuhat.helsinki.fi/ws/portalfiles/portal/134581230/remotesensing_12_01070_v2_1_.pdf

Abstract

Interest in drone solutions in forestry applications is growing. Using drones, datasets can be captured flexibly and at high spatial and temporal resolutions when needed. In forestry applications, fundamental tasks include the detection of individual trees, tree species classification, biomass estimation, etc. Deep neural networks (DNN) have shown superior results when comparing with conventional machine learning methods such as multi-layer perceptron (MLP) in cases of huge input data. The objective of this research is to investigate 3D convolutional neural networks (3D-CNN) to classify three major tree species in a boreal forest: pine, spruce, and birch. The proposed 3D-CNN models were employed to classify tree species in a test site in Finland. The classifiers were trained with a dataset of 3039 manually labelled trees. Then the accuracies were assessed by employing independent datasets of 803 records. To find the most efficient set of feature combination, we compare the performances of 3D-CNN models trained with hyperspectral (HS) channels, Red-Green-Blue (RGB) channels, and canopy height model (CHM), separately and combined. It is demonstrated that the proposed 3D-CNN model with RGB and HS layers produces the highest classification accuracy. The producer accuracy of the best 3D-CNN classifier on the test dataset were 99.6%, 94.8%, and 97.4% for pines, spruces, and birches, respectively. The best 3D-CNN classifier produced ~5% better classification accuracy than the MLP with all layers. Our results suggest that the proposed method provides excellent classification results with acceptable performance metrics for HS datasets. Our results show that pine class was detectable in most layers. Spruce was most detectable in RGB data, while birch was most detectable in the HS layers. Furthermore, the RGB datasets provide acceptable results for many low-accuracy applications.

Keywords: unmanned aerial vehicles; remote sensing; forest mensuration; tree species; spectral imaging; machine learning; neural networks (information technology)

Free keywords: deep learning; drone imagery; hyperspectral image classification; tree species classification; 3D convolutional neural networks

Contributing organizations

Ministry reporting: Yes

Reporting Year: 2020

JUFO rating: 1