Weed control is one of the biggest challenges in organic farms or nature reserve areas where mass spraying is prohibited. Recent advancements in remote sensing and airborne technologies provide a fast and efficient means to support environmental monitoring and management, allowing early detection of invasive species. However, in order to perform weed classification, current studies mostly relied on object-based image analysis (OBIA) and proprietary software which required substantial human inputs. This paper proposes an open-source workflow for automated weed mapping using a commercially available unmanned aerial vehicle (UAV). The UAV was flown at a low altitude between 10 m and 20 m, and collected truecolour RGB imagery over a weed-infested nature reserve. The aim of this study is to develop a repeatable and robust system for early weed detection, with minimum human intervention, for classification of Rumex obtusifolius (R. obtusifolius). Preliminary results of the proposed workflow achieved an overall accuracy of 92.1% with an F1 score of 78.7%. The approach also demonstrated the capability to map R. obtusifolius in datasets collected at various flight altitudes, camera settings and light conditions. This shows the potential to perform semi-or fully automated early weed detection system in grasslands using UAV-imagery.
Rumex obtusifolius (R. obtusifolius) is one of the most common non-cultivated weed in European grasslands. Its broad-leaved and wide-spread nature make this weed competitive with the native pasture species reducing grass yield (van Evert et al. 2010), while its oxalic acid content makes this species poisonous for livestock if large doses are consumed (Hejduk and Doležal 2004). Therefore, early removal is preferred especially in organic dairy farms or conservation areas where mass spraying is prohibited. Remote sensing and airborne technologies offer fast and efficient support in environmental monitoring allowing early detection of invasive species, yet current studies mostly rely on object-based image analysis (OBIA) and proprietary software to perform weed classification that require substantial human inputs. In this work, an open source workflow for automatic weed detection using unmanned aerial vehicle (UAV) RGB-imagery of native grassland had been developed using deep learning techniques, based on a previously developed OBIA approach (Lam et al. 2019). During the study, DJI Phantom 3 and 4 Pro were used for data acquisition throughout the vegetation period in 2018 and early 2019 at a nature conversation area in North Rhine-Westphalia, Germany. Images were processed using OpenDroneMap to produce orthomosaics. OBIA methods were then performed using Python and QGIS to assist the data labelling process for training a convolutional neural network (CNN), which was later used as an image classifier. Preliminary results of the proposed workflow achieved an overall accuracy of 93.8% and had demonstrated the capability in mapping R. obtusifolius in datasets collected at various flight altitudes, camera settings and light conditions. This shows the potential of developing a repeatable and robust system for semi-or fully-automated early weed detection in grassland using UAV-imagery.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.