Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding
- Author
- Axel Willekens (UGent) , Bert Callens, Francis wyffels (UGent) , Jan Pieters (UGent) and Simon R. Cool
- Organization
- Abstract
- Mechanical weeding is an important part of integrated weed management. It destroys weeds between (interrow) and in (intrarow) crop rows. Preventing crop damage requires precise detection and tracking of the plants. In this work, a detection and tracking algorithm was developed and integrated on an intrarow hoeing prototype. The algorithm was developed and validated on 12 rows of 950 cauliflower plants. Therefore, a methodology was provided to automatically generate a label based on the crop plants’ Global Navigation Satellite System (GNSS) position during data collection with a robot platform. A CenterNet architecture was adjusted for plant centre detection by comparing different encoder networks and selecting the optimal hyperparameters. The monocular camera projection error of the plant centre detections in pixel to 3D coordinates was evaluated and used in a position- and velocity-based tracking algorithm to determine the timing for intrarow hoeing knife actuation. A dataset of 53k labelled images was created. The best CenterNet model resulted in an F1 score on the test set of 0.986 for detecting cauliflower centres. The position tracking had an average variation of 1.62 cm. Velocity tracking had a standard deviation of 0.008 with respect to the robot’s operational target velocity. Overall, the entire integration showed effective actuation of the prototype in field conditions. Only one false positive detection occurred during operation in two test rows of 135 cauliflowers.
- Keywords
- Robotic intrarow weeding, Cauliflower, GNSS labels, CenterNet detection, Mono camera localisation, CLASSIFICATION, LOCALIZATION, AGRICULTURE, LOSSES
Downloads
-
paper cauliflower center detection R1.pdf
- full text (Accepted manuscript)
- |
- open access
- |
- |
- 62.94 MB
-
(...).pdf
- full text (Published version)
- |
- UGent only
- |
- |
- 6.14 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01JKAMEAZK5FRM8RPZTC4TP04Y
- MLA
- Willekens, Axel, et al. “Cauliflower Centre Detection and 3-Dimensional Tracking for Robotic Intrarow Weeding.” PRECISION AGRICULTURE, vol. 26, no. 2, 2025, doi:10.1007/s11119-025-10227-3.
- APA
- Willekens, A., Callens, B., wyffels, F., Pieters, J., & Cool, S. R. (2025). Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding. PRECISION AGRICULTURE, 26(2). https://doi.org/10.1007/s11119-025-10227-3
- Chicago author-date
- Willekens, Axel, Bert Callens, Francis wyffels, Jan Pieters, and Simon R. Cool. 2025. “Cauliflower Centre Detection and 3-Dimensional Tracking for Robotic Intrarow Weeding.” PRECISION AGRICULTURE 26 (2). https://doi.org/10.1007/s11119-025-10227-3.
- Chicago author-date (all authors)
- Willekens, Axel, Bert Callens, Francis wyffels, Jan Pieters, and Simon R. Cool. 2025. “Cauliflower Centre Detection and 3-Dimensional Tracking for Robotic Intrarow Weeding.” PRECISION AGRICULTURE 26 (2). doi:10.1007/s11119-025-10227-3.
- Vancouver
- 1.Willekens A, Callens B, wyffels F, Pieters J, Cool SR. Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding. PRECISION AGRICULTURE. 2025;26(2).
- IEEE
- [1]A. Willekens, B. Callens, F. wyffels, J. Pieters, and S. R. Cool, “Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding,” PRECISION AGRICULTURE, vol. 26, no. 2, 2025.
@article{01JKAMEAZK5FRM8RPZTC4TP04Y,
abstract = {{Mechanical weeding is an important part of integrated weed management. It destroys weeds between (interrow) and in (intrarow) crop rows. Preventing crop damage requires precise detection and tracking of the plants. In this work, a detection and tracking algorithm was developed and integrated on an intrarow hoeing prototype. The algorithm was developed and validated on 12 rows of 950 cauliflower plants. Therefore, a methodology was provided to automatically generate a label based on the crop plants’ Global Navigation Satellite System (GNSS) position during data collection with a robot platform. A CenterNet architecture was adjusted for plant centre detection by comparing different encoder networks and selecting the optimal hyperparameters. The monocular camera projection error of the plant centre detections in pixel to 3D coordinates was evaluated and used in a position- and velocity-based tracking algorithm to determine the timing for intrarow hoeing knife actuation. A dataset of 53k labelled images was created. The best CenterNet model resulted in an F1 score on the test set of 0.986 for detecting cauliflower centres. The position tracking had an average variation of 1.62 cm. Velocity tracking had a standard deviation of 0.008 with respect to the robot’s operational target velocity. Overall, the entire integration showed effective actuation of the prototype in field conditions. Only one false positive detection occurred during operation in two test rows of 135 cauliflowers.}},
articleno = {{29}},
author = {{Willekens, Axel and Callens, Bert and wyffels, Francis and Pieters, Jan and Cool, Simon R.}},
issn = {{1385-2256}},
journal = {{PRECISION AGRICULTURE}},
keywords = {{Robotic intrarow weeding,Cauliflower,GNSS labels,CenterNet detection,Mono camera localisation,CLASSIFICATION,LOCALIZATION,AGRICULTURE,LOSSES}},
language = {{eng}},
number = {{2}},
pages = {{26}},
title = {{Cauliflower centre detection and 3-dimensional tracking for robotic intrarow weeding}},
url = {{http://doi.org/10.1007/s11119-025-10227-3}},
volume = {{26}},
year = {{2025}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: