Deep learning‐based weed detection for precision herbicide application in turf
Pest Management Science · 2025
DOI: 10.1002/ps.8728 ↗Kurz gefasst
Fachartikel in der Zeitschrift Pest Management Science, 2025. Verfasst von Xiaojun Jin und 7 weiteren Autor:innen. Beteiligte Einrichtung: College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing, China. Hauptthemen laut MEDLINE: Herbicides, Weed Control, Deep Learning, Plant Weeds. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Weifang Science and Technology Development Plan Project, Key R&D Program of Shandong Province, China.
Abstract (Originalsprache)
BACKGROUND Precision weed mapping in turf according to its susceptibility to selective herbicides allows the smart sprayer to spot‐spray the most pertinent herbicides onto the susceptible weeds. The objective of this study was to evaluate the feasibility of implementing herbicide susceptibility‐based weed mapping using deep convolutional neural networks (DCNNs) to facilitate targeted and efficient herbicide applications. Additionally, applying path‐planning algorithms to weed mapping data to guide the spraying nozzle ensures minimal travel paths for herbicide application. RESULTS DenseNet achieved high precision, recall, overall accuracy, and F 1 score values for all categories of herbicides and no herbicides, with F 1 scores ranging from 0.996 to 0.999 in the validation dataset and from 0.992 to 0.997 in the testing dataset. The average accuracies attained by DenseNet, GoogLeNet and ResNet were 0.9985, 0.9953 and 0.9980, respectively. By considering both accuracy and computational efficiency, the ResNet model was identified as the most effective among the models compared to weed detection. The performance of the Christofides, Greedy and 2‐opt algorithms in optimizing path planning for single or dual spraying nozzles was compared and analyzed. The Greedy algorithm proved the most efficient in optimizing the nozzle's trajectory. CONCLUSION Implementing herbicide susceptibility‐based weed mapping facilitates targeted herbicide application by directing the nozzle to the grid cells containing the weeds susceptible to the herbicides. Moreover, the strategic integration of herbicide susceptibility‐based weed mapping with optimized path planning for the spraying mechanism can be adeptly implemented on smart sprayers, which could effectively reduce the herbicide input. © 2025 Society of Chemical Industry.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing, China.
- Peking University Institute of Advanced Agricultural Sciences, Shandong Laboratory of Advanced Agricultural Sciences in Weifang, Shandong, China.
ORCID-Profile: Xiaojun Jin, Jinglin Lei
Förderung
- National Natural Science Foundation of China
- Weifang Science and Technology Development Plan Project
- Key R&D Program of Shandong Province, China
Zitieren
APA
Jin, X., Zhao, H., Kong, X., Han, K., Lei, J., Zu, Q., Chen, Y., & Yu, J. (2025). Deep learning‐based weed detection for precision herbicide application in turf. Pest Management Science. https://doi.org/10.1002/ps.8728
Vancouver
Jin X, Zhao H, Kong X, Han K, Lei J, Zu Q, et al. Deep learning‐based weed detection for precision herbicide application in turf. Pest Management Science. 2025. doi:10.1002/ps.8728
BibTeX
@article{Jin2025,
author = {Jin, Xiaojun and Zhao, Hua and Kong, Xiaotong and Han, Kang and Lei, Jinglin and Zu, Qiuyu and Chen, Yong and Yu, Jialin},
title = {Deep learning‐based weed detection for precision herbicide application in turf},
journal = {Pest Management Science},
year = {2025},
doi = {10.1002/ps.8728}
}
EUVIMED