article · Frontiers in Plant Science
Recent advances in computer vision and machine learning enable automated plant monitoring using drone imagery. This research establishes a practical workflow using the YOLOv5 object detection algorithm and high-resolution aerial photographs to detect and count maize plants across different growth stages. To reduce the manual effort required for data preparation, a semi-automated annotation method using the Segment Anything Model was incorporated. The resulting system successfully identified maize plants at both the three-leaf and seven-leaf developmental stages, achieving strong precision even in challenging field conditions such as overgrown weeds, leaf occlusion, and blurry images. Adding image-rotation data augmentation and tuning noise weights further boosted detection performance. The findings demonstrate a viable framework for characterising crop growth and automating plant counts under realistic agricultural settings.
Accurate plant counting is essential for assessing crop emergence, estimating yield, and managing field inputs effectively. Traditional manual counting is labour-intensive and slow across large farming areas. By demonstrating that drone images and lightweight computer vision models can reliably count crops amidst weeds and visual clutter, this approach offers an efficient path toward automated, field-scale crop monitoring.
This workflow is relevant to developers of precision agriculture software, drone service providers, and agricultural managers seeking automated crop counting tools. The system demonstrates applied and tested status under realistic field conditions, such as weed-heavy environments. Moving toward commercial deployment would require integrating the pipeline into operational agricultural analytics platforms and testing across a broader variety of field layouts and crop types.
AI-generated from the published abstract. Always read the original work before citing.
In recent years, computer vision (CV) has made enormous progress and is providing great possibilities in analyzing images for object detection, especially with the application of machine learning (ML). Unmanned Aerial Vehicle (UAV) based high-resolution images allow to apply CV and ML methods for the detection of plants or their organs of interest. Thus, this study presents a practical workflow based on the You Only Look Once version 5 (YOLOv5) and UAV images to detect maize plants for counting their numbers in contrasting development stages, including the application of a semi-auto-labeling method based on the Segment Anything Model (SAM) to reduce the burden of labeling. Results showed that the trained model achieved a mean average precision (mAP@0.5) of 0.828 and 0.863 for the 3-leaf stage and 7-leaf stage, respectively. YOLOv5 achieved the best performance under the conditions of overgrown weeds, leaf occlusion, and blurry images, suggesting that YOLOv5 plays a practical role in obtaining excellent performance under realistic field conditions. Furthermore, introducing image-rotation augmentation and low noise weight enhanced model accuracy, with an increase of 0.024 and 0.016 mAP@0.5, respectively, compared to the original model of the 3-leaf stage. This work provides a practical reference for applying lightweight ML and deep learning methods to UAV images for automated object detection and characterization of plant growth under realistic environments.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3389/fpls.2023.1274813
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
