• search hit 1 of 2
Back to Result List

Indicator plant species detection in grassland using EfficientDet object detector

  • Extensively used grasslands (meadows and pastures) are ecologically valuable areas in the agricultural landscape and part of the multifunctional agriculture. In Germany, the quality of these grasslands is assessed based on the occurrence of certain plant species known as indicator or character species, with indicators being defined at regional level. Therefore, the recognition of these indicators on a spatial level is a prerequisite for monitoring grassland biodiversity. The identification of indicator species for the status quo of grassland using traditional methods was found to be challenging and tedious. Deep learning-algorithms applied to high-resolution UAV imagery could be the key solution, where UAV with remote sensors can map a large area of grassland in comparison to manual or ground mapping methods and deep learning-algorithms can automate the detection process. In this research work, we use an EfficientDet based algorithm to train an object detection model capable of recognizing indicators on RGB data. The experimentalExtensively used grasslands (meadows and pastures) are ecologically valuable areas in the agricultural landscape and part of the multifunctional agriculture. In Germany, the quality of these grasslands is assessed based on the occurrence of certain plant species known as indicator or character species, with indicators being defined at regional level. Therefore, the recognition of these indicators on a spatial level is a prerequisite for monitoring grassland biodiversity. The identification of indicator species for the status quo of grassland using traditional methods was found to be challenging and tedious. Deep learning-algorithms applied to high-resolution UAV imagery could be the key solution, where UAV with remote sensors can map a large area of grassland in comparison to manual or ground mapping methods and deep learning-algorithms can automate the detection process. In this research work, we use an EfficientDet based algorithm to train an object detection model capable of recognizing indicators on RGB data. The experimental results show that this approach is very promising in contrast to the difficult and time-consuming manual recognition methods. The model was trained with the momentum-SGD optimizer with a momentum value of 0.9 and a learning rate of 0.0001. The model was trained and tested on 1200 images and achieves 45.7 AP (and 85.7 AP50) on test data set. The dataset includes images of four distinct indicator plant species: Armeria maritima, Campanula patula, Cirsium oleraceum, and Daucus carotashow moreshow less

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar Statistics
Metadaten
Author:Deepak Hanike Basavegowda, Paul Mosebach, Inga Schleip, Cornelia Weltzien
URN:urn:nbn:de:kobv:eb1-opus-6656
ISBN:978-3-88579-711-1
ISSN:1617-5468
Parent Title (German):42. GIL-Jahrestagung, Künstliche Intelligenz in der Agrar- und Ernährungswirtschaft
Publisher:Gesellschaft für Informatik e.V.
Place of publication:Bonn
Document Type:Conference Proceeding
Language:English
Year of Completion:2022
Publishing Institution:Hochschule für nachhaltige Entwicklung Eberswalde
Release Date:2023/09/13
Tag:digital agriculture; biodiversity in grassland; HNV farming; deep learning; object detection
First Page:57
Last Page:62
Institutions / Departments:Fachbereich Landschaftsnutzung und Naturschutz
University Bibliography:University Bibliography
Zweitveröffentlichung
Peer-Review / Referiert
Licence (German):License LogoUrheberrechtsschutz
Einverstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.