Identifying plant species in kettle holes using UAV images and deep learning techniques

Publikation: Beitrag in FachzeitschriftForschungsartikelBeigetragenBegutachtung


  • José Augusto Correa Martins - (Autor:in)
  • José Marcato Junior - (Autor:in)
  • Marlene Pätzig - (Autor:in)
  • Diego André Sant'Ana - (Autor:in)
  • Hemerson Pistori - (Autor:in)
  • Veraldo Liesenberg - (Autor:in)
  • Anette Eltner - , Juniorprofessur für Geosensorsysteme (TT) (Autor:in)


The use of uncrewed aerial vehicle to map the environment increased significantly in the last decade enabling a finer assessment of the land cover. However, creating accurate maps of the environment is still a complex and costly task. Deep learning (DL) is a new generation of artificial neural network research that, combined with remote sensing techniques, allows a refined understanding of our environment and can help to solve challenging land cover mapping issues. This research focuses on the vegetation segmentation of kettle holes. Kettle holes are small, pond-like, depressional wetlands. Quantifying the vegetation present in this environment is essential to assess the biodiversity and the health of the ecosystem. A machine learning workflow has been developed, integrating a superpixel segmentation algorithm to build a robust dataset, which is followed by a set of DL architectures to classify 10 plant classes present in kettle holes. The best architecture for this task was Xception, which achieved an average F1-score of 85% in the segmentation of the species. The application of solely 318 samples per class enabled a successful mapping in the complex wetland environment, indicating an important direction for future health assessments in such landscapes.


Seiten (von - bis)1-16
FachzeitschriftRemote Sensing in Ecology and Conservation
PublikationsstatusVeröffentlicht - Feb. 2023

Externe IDs

Scopus 85135857394
WOS 000839367100001
Mendeley 88a78049-8c51-3171-9bf4-0e90174c0a3a
unpaywall 10.1002/rse2.291


Ziele für nachhaltige Entwicklung


  • Deep learning, image segmentation, plant species segmentation, superpixels, uncrewed aerial vehicle, wetland, Image segmentation, Uncrewed aerial vehicle, Plant species segmentation, Superpixels, Wetland