Key points are not available for this paper at this time.
This study presents a novel Unmanned Aerial Vehicle (UAV)-based approach for detecting pigweed in soybean (Glycine max) fields using a combination of deep learning and advanced image processing techniques. A custom, high-resolution dataset comprising RGB and multispectral images was collected from USDA operated fields and manually annotated to target pigweed detection. Beyond incorporating YOLOv8 variants for real-time weed classification, this research integrates a comprehensive image processing pipeline, incorporating global thresholding, k-means clustering, 3D surface mapping, and spectral signature analysis to enhance interpretability and detection accuracy. A comparative evaluation of YOLOv8 nano, small, medium, and large models was performed to identify the most practical model for deployment in precision agriculture. The YOLOv8 nano model emerged as the most balanced in terms of precision (75.6%), recall (81.7%), and mAP@0.5 (81.6%), demonstrating effective weed detection performance under real field conditions. Also, quadrant-level weed coverage and spatial heatmaps were generated to support targeted intervention. This work advances the current state of UAV based weed detection by providing a field-ready, explainable, and resource-efficient solution, contributing to sustainable farming and data-driven weed management practices.
Borah et al. (Wed,) studied this question.