Open AccessIdentifying plant species in kettle holes using UAV images and deep learning techniques
Author(s) -
Correa Martins José Augusto,
Marcato Junior José,
Pätzig Marlene,
Sant'Ana Diego André,
Pistori Hemerson,
Liesenberg Veraldo,
Eltner Anette
Publication year - 2023
Publication title -
remote sensing in ecology and conservation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.191
H-Index - 21
ISSN - 2056-3485
DOI - 10.1002/rse2.291
Subject(s) - kettle (birds) , workflow , wetland , segmentation , computer science , artificial intelligence , task (project management) , vegetation (pathology) , land cover , deep learning , set (abstract data type) , cluster analysis , artificial neural network , environmental science , machine learning , remote sensing , geography , land use , engineering , ecology , civil engineering , systems engineering , biology , database , medicine , pathology , programming language
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 F 1‐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.