Abstract
Weeds are a persistent problem on sod farms, and herbicides to control different weed species are one of the largest chemical inputs. Recent advances in unmanned aerial systems (UAS) and artificial intelligence provide opportunities for weed mapping on sod farms. This study investigates the weed type composition and area through both ground and UAS-based weed surveys and trains a convolutional neural network (CNN) for identifying and mapping weeds in sod fields using UAS-based imagery and a high-level application programming interface (API) implementation (Fastai) of the PyTorch deep learning library. The performance of the CNN was overall similar to, and in some classes (broadleaf and spurge) better than, human eyes indicated by the metric recall. In general, the CNN detected broadleaf, grass weeds, spurge, sedge, and no weeds at a precision between 0.68 and 0.87, 0.57 and 0.82, 0.68 and 0.83, 0.66 and 0.90, and 0.80 and 0.88, respectively, when using UAS images at 0.57 cm–1.28 cm pixel–1 resolution. Recall ranges for the five classes were 0.78–0.93, 0.65–0.87, 0.82–0.93, 0.52–0.79, and 0.94–0.99. Additionally, this study demonstrates that a CNN can achieve precision and recall above 0.9 at detecting different types of weeds during turf establishment when the weeds are mature. The CNN is limited by the image resolution, and more than one model may be needed in practice to improve the overall performance of weed mapping.
Highlights
Weeds are a persistent problem on sod farms
This study included the survey of several sod production fields for broadleaf, grass weeds, spurge, and sedge weed-type composition and areas of infestation, both from the ground level and using UAS, demonstrating the potential of herbicide savings if sitespecific weed management is properly adopted
It was demonstrated that the CNN can achieve precision and recall above 0.9 for detecting
Summary
Weeds are a persistent problem on sod farms. Herbicides to control different weed species are one of the largest chemical inputs (Satterthwaite et al, 2009; Wojciech and Landry, 2009; Yi, 2012) and often their control requires multiple applications throughout the growing season. Regulations limiting the broadcast application of certain chemicals in sod production (USEPA, 2009), due to concerns about the UAS Weed Mapping in Turfgrass environmental impacts of the herbicide, create difficulty in effectively controlling weeds. Aside from the environmental cost of herbicides, there are significant financial costs in purchasing the herbicide and the labor and fuel used in application. Sitespecific weed management, such as applying herbicides only where the weeds are located, instead of whole-field broadcast applications would significantly reduce herbicide use, thereby improving economic and environmental sustainability in sod production. The ability to quickly identify and respond to areas with weed issues is an attractive proposition for both sod growers and inspection agencies
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call