Evaluation of convolutional neural networks for herbicide susceptibility-based weed detection in turf

Xiaojun Jin; Teng Liu; Patrick E McCullough; Yong Chen; Jialin Yu

doi:10.3389/fpls.2023.1096802

Evaluation of convolutional neural networks for herbicide susceptibility-based weed detection in turf

Front Plant Sci. 2023 Feb 1:14:1096802. doi: 10.3389/fpls.2023.1096802. eCollection 2023.

Authors

Xiaojun Jin^{1

2}, Teng Liu², Patrick E McCullough³, Yong Chen¹, Jialin Yu²

Affiliations

¹ College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing, Jiangsu, China.
² Peking University Institute of Advanced Agricultural Sciences / Shandong Laboratory of Advanced Agricultural Sciences at Weifang, Weifang, Shandong, China.
³ Department of Crop and Soil Sciences, University of Georgia, Griffin, GA, United States.

Abstract

Deep learning methods for weed detection typically focus on distinguishing weed species, but a variety of weed species with comparable plant morphological characteristics may be found in turfgrass. Thus, it is difficult for deep learning models to detect and distinguish every weed species with high accuracy. Training convolutional neural networks for detecting weeds susceptible to herbicides can offer a new strategy for implementing site-specific weed detection in turf. DenseNet, EfficientNet-v2, and ResNet showed high F₁ scores (≥0.986) and MCC values (≥0.984) to detect and distinguish the sub-images containing dollarweed, goosegrass, old world diamond-flower, purple nutsedge, or Virginia buttonweed growing in bermudagrass turf. However, they failed to reliably detect crabgrass and tropical signalgrass due to the similarity in plant morphology. When training the convolutional neural networks for detecting and distinguishing the sub-images containing weeds susceptible to ACCase-inhibitors, weeds susceptible to ALS-inhibitors, or weeds susceptible to synthetic auxin herbicides, all neural networks evaluated in this study achieved excellent F₁ scores (≥0.995) and MCC values (≥0.994) in the validation and testing datasets. ResNet demonstrated the fastest inference rate and outperformed the other convolutional neural networks on detection efficiency, while the slow inference of EfficientNet-v2 may limit its potential applications. Grouping different weed species growing in turf according to their susceptibility to herbicides and detecting and distinguishing weeds by herbicide categories enables the implementation of herbicide susceptibility-based precision herbicide application. We conclude that the proposed method is an effective strategy for site-specific weed detection in turf, which can be employed in a smart sprayer to achieve precision herbicide spraying.

Keywords: convolutional neural networks; deep learning; herbicide susceptibility; precision herbicide application; weed detection.

Grants and funding

This work was supported by the National Natural Science Foundation of China (Grant No. 32072498) and the Postgraduate Research & Practice Innovation Program of Jiangsu Province (Grant No. KYCX22_1051).