EUVIMEDEUVIMED entdecken
Crossref

Synthetic data augmentation by diffusion probabilistic models to enhance weed recognition

Dong Chen, Xinda Qi, Yu Zheng, Yuzhen Lu, Yanbo Huang, Zhaojian Li

Computers and Electronics in Agriculture · 2024-01 · 216 · 108517

DOI: 10.1016/j.compag.2023.108517 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers and Electronics in Agriculture, 2024-01. Verfasst von Dong Chen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 77-mal zitiert (Crossref) und nennt 67 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Weed management plays an important role in crop yield and quality protection. Conventional weed control methods largely rely on herbicide application, which incurs significant management costs and poses environmental hazards to the environments and human health. Machine vision-based automated weeding has gained increasing attention for sustainable weed management through weed identification and site-specific treatments. However, it remains a challenging task to reliably recognize weeds in variable field conditions, which is largely because of the difficulty curating large-scale, expert-labeled weed image datasets collected from diverse field conditions. Data augmentation methods, including traditional geometric/color transformations and more advanced generative adversarial networks (GANs) can alleviate data collection and labeling efforts by algorithmically expanding the sale of datasets. Recently, diffusion models have emerged in the field of image synthesis, providing a new means for augmenting image datasets to power machine vision systems. This study presents the first investigation of the efficacy of diffusion models for generating weed images to enhance weed identification. Experiments on a 15-class large weed data (CottonWeedID15), showed that diffusion models yielded achieved the best trade-off between sample fidelity and diversity and obtained the highest Fréchet Inception Distance, compared to GANs (BigGAN, StyleGAN2, StyleGAN3), the dataset expanded with synthetic weed images improved (up to 1.20%, 1.23%, and 2.30% on the accuracy, precision, and recall, respectively) weed classification accuracy by four deep learning models (i.e., VGG16, Inception-v3, Inception-v3, and ResNet50). Models trained using only 10% of real images with the remainder being synthetic data resulted in testing accuracy exceeding 94%.

Zitieren

APA

Chen, D., Qi, X., Zheng, Y., Lu, Y., Huang, Y., & Li, Z. (2024). Synthetic data augmentation by diffusion probabilistic models to enhance weed recognition. Computers and Electronics in Agriculture, 216, 108517. https://doi.org/10.1016/j.compag.2023.108517

Vancouver

Chen D, Qi X, Zheng Y, Lu Y, Huang Y, Li Z. Synthetic data augmentation by diffusion probabilistic models to enhance weed recognition. Computers and Electronics in Agriculture. 2024;216:108517. doi:10.1016/j.compag.2023.108517

BibTeX
@article{Chen2024,
  author = {Chen, Dong and Qi, Xinda and Zheng, Yu and Lu, Yuzhen and Huang, Yanbo and Li, Zhaojian},
  title = {Synthetic data augmentation by diffusion probabilistic models to enhance weed recognition},
  journal = {Computers and Electronics in Agriculture},
  year = {2024},
  volume = {216},
  pages = {108517},
  doi = {10.1016/j.compag.2023.108517}
}
Kostenlos starten →