WE3DS: An RGB-D Image Dataset for Semantic Segmentation in Agriculture
Sensors · 01.03.2023 · 23 (5) · 2713
DOI: 10.3390/s23052713 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 01.03.2023. Verfasst von Florian Kitzler und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Sustainable Agricultural Systems, Institute of Agricultural Engineering, University of Natural Resources and Life Sciences Vienna, Pete…. Die Arbeit wurde bisher 33-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Government of Lower Austria, Government of Lower Austria and the private foundation Forum Morgen.
Abstract (Originalsprache)
Smart farming (SF) applications rely on robust and accurate computer vision systems. An important computer vision task in agriculture is semantic segmentation, which aims to classify each pixel of an image and can be used for selective weed removal. State-of-the-art implementations use convolutional neural networks (CNN) that are trained on large image datasets. In agriculture, publicly available RGB image datasets are scarce and often lack detailed ground-truth information. In contrast to agriculture, other research areas feature RGB-D datasets that combine color (RGB) with additional distance (D) information. Such results show that including distance as an additional modality can improve model performance further. Therefore, we introduce WE3DS as the first RGB-D image dataset for multi-class plant species semantic segmentation in crop farming. It contains 2568 RGB-D images (color image and distance map) and corresponding hand-annotated ground-truth masks. Images were taken under natural light conditions using an RGB-D sensor consisting of two RGB cameras in a stereo setup. Further, we provide a benchmark for RGB-D semantic segmentation on the WE3DS dataset and compare it with a solely RGB-based model. Our trained models achieve up to 70.7% mean Intersection over Union (mIoU) for discriminating between soil, seven crop species, and ten weed species. Finally, our work confirms the finding that additional distance information improves segmentation quality.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Sustainable Agricultural Systems, Institute of Agricultural Engineering, University of Natural Resources and Life Sciences Vienna, Peter-Jordan-Straße 82, 1190 Vienna, Austria.
- Department of Crop Sciences, Institute of Agronomy, University of Natural Resources and Life Sciences Vienna, Konrad Lorenz-Straße 24, 3430 Tulln an der Donau, Austria.
ORCID-Profile: Florian Kitzler, Norbert Barta, Reinhard W Neugschwandtner, Andreas Gronauer, Viktoria Motsch
Förderung
- Government of Lower Austria
- Government of Lower Austria and the private foundation Forum Morgen
Zitieren
APA
Kitzler, F., Barta, N., Neugschwandtner, R. W., Gronauer, A., & Motsch, V. (2023). WE3DS: An RGB-D Image Dataset for Semantic Segmentation in Agriculture. Sensors, 23(5), 2713. https://doi.org/10.3390/s23052713
Vancouver
Kitzler F, Barta N, Neugschwandtner RW, Gronauer A, Motsch V. WE3DS: An RGB-D Image Dataset for Semantic Segmentation in Agriculture. Sensors. 2023;23(5):2713. doi:10.3390/s23052713
BibTeX
@article{Kitzler2023,
author = {Kitzler, Florian and Barta, Norbert and Neugschwandtner, Reinhard W and Gronauer, Andreas and Motsch, Viktoria},
title = {WE3DS: An RGB-D Image Dataset for Semantic Segmentation in Agriculture},
journal = {Sensors},
year = {2023},
volume = {23},
number = {5},
pages = {2713},
doi = {10.3390/s23052713}
}
EUVIMED