OpenWeedLocator (OWL): an open-source, low-cost device for fallow weed detection

Abstract

The use of a fallow phase is an important tool for maximizing crop yield potential in moisture limited agricultural environments, with a focus on removing weeds to optimize fallow efficiency. Repeated whole field herbicide treatments to control low-density weed populations is expensive and wasteful. Site-specific herbicide applications to low-density fallow weed populations is currently facilitated by proprietary, sensor-based spray booms. The use of image analysis for fallow weed detection is an opportunity to develop a system with potential for in-crop weed recognition. Here we present OpenWeedLocator (OWL), an open-source, low-cost and image-based device for fallow weed detection that improves accessibility to this technology for the weed control community. A comprehensive GitHub repository was developed, promoting community engagement with site-specific weed control methods. Validation of OWL as a low-cost tool was achieved using four, existing colour-based algorithms over seven fallow fields in New South Wales, Australia. The four algorithms were similarly effective in detecting weeds with average precision of 79% and recall of 52%. In individual transects up to 92% precision and 74% recall indicate the performance potential of OWL in fallow fields. OWL represents an opportunity to redefine the approach to weed detection by enabling community-driven technology development in agriculture.

Advancements in precision control in intra-row weeding: A comprehensive review

“…CED-Net's efficiency with fewer parameters results in shorter training and inference times, and it outperforms or matches other state-of-the-art networks across multiple evaluation metrics, while requiring only 1/3 to 1/5 fewer parameters than comparable models like U-Net and SegNet. Lastly, Coleman et al (2022) [22] integrated mechatronic systems with AI-driven algorithms, leading to more effective and environmentally friendly herbicide applications in maize fields. These developments collectively underscore the dynamic and evolving landscape of non-image sensor systems for intra-row weeding.…”

A Review on the Use of Open Source Technologies and Soft-Ware Applied to Precision Agriculture Practices

“…According to the performed experiments, the authors claim that the differential location accuracy is such that driving the robot to revisit specific plants is possible for seed spacings up to 139 mm. Coleman (2022) developed the OpenWeedLocator (OWL) project with the aim of maximizing crop yield potential in agricultural environments with limited humidity and a focus on weed removal (Table III, item 3). The system integrates weed detection on a Raspberry Pi with a relay control board that mounts on robots and vehicles for timely spraying.…”

More eyes on the prize: open-source data, software and hardware for advancing plant science through collaboration

“…Robotic gantry systems such as Farmbot ( https://github.com/FarmBot ) could serve as low cost and customisable systems for data collection in fixed semi-field settings. For data collection, the recently released OpenWeedLocator (OWL) project ( Coleman et al , 2022 ) provides a method for image collection and integration with colour-based detection algorithms ( https://github.com/geezacoleman/OpenWeedLocator ). While developed with weed detection in mind, the tool could be adapted for image data collection.…”