KIT | KIT-Bibliothek | Impressum | Datenschutz

Mapping forest tree species in high resolution UAV-based RGB-imagery by means of convolutional neural networks

Schiefer, Felix; Kattenborn, Teja; Frick, Annett; Frey, Julian; Schall, Peter; Koch, Barbara; Schmidtlein, Sebastian

The use of unmanned aerial vehicles (UAVs) in vegetation remote sensing allows a time-flexible and cost-effective acquisition of very high-resolution imagery. Still, current methods for the mapping of forest tree species do not exploit the respective, rich spatial information. Here, we assessed the potential of convolutional neural networks (CNNs) and very high-resolution RGB imagery from UAVs for the mapping of tree species in temperate forests. We used multicopter UAVs to obtain very high-resolution (<2 cm) RGB imagery over 51 ha of temperate forests in the Southern Black Forest region, and the Hainich National Park in Germany. To fully harness the end-to-end learning capabilities of CNNs, we used a semantic segmentation approach (U-net) that concurrently segments and classifies tree species from imagery. With a diverse dataset in terms of study areas, site conditions, illumination properties, and phenology, we accurately mapped nine tree species, three genus-level classes, deadwood, and forest floor (mean F1-score 0.73). A larger tile size during CNN training negatively affected the model accuracies for underrepresented classes. ... mehr

DOI: 10.1016/j.isprsjprs.2020.10.015
Zugehörige Institution(en) am KIT Institut für Geographie und Geoökologie (IFGG)
Publikationstyp Zeitschriftenaufsatz
Publikationsdatum 01.12.2020
Sprache Englisch
Identifikator ISSN: 0924-2716
KITopen-ID: 1000125945
Erschienen in ISPRS journal of photogrammetry and remote sensing
Band 170
Seiten 205–215
Vorab online veröffentlicht am 03.11.2020
Schlagwörter Deep learning; Forest inventory; Convolutional neural networks; Tree species classification; Unmanned aerial systems; Temperate forests
Nachgewiesen in Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page