EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning Models for Detection and Severity Assessment of Cercospora Leaf Spot (Cercospora capsici) in Chili Peppers Under Natural Conditions

Douglas Vieira Leite iD, Alisson Vasconcelos de Brito iD, Gregorio Guirada Faccioli iD, Gustavo Haddad Souza Vieira iD

Plants · 01.07.2025 · 14 (13) · 2011

DOI: 10.3390/plants14132011 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Plants, online erschienen am 01.07.2025. Verfasst von Douglas Vieira Leite und 3 weiteren Autor:innen. Beteiligte Einrichtung: EMEC, Sergipe Educational, Technology and Scientific Institute, Lourival Batista, Lourival Batista Highway s/n, Lagarto 49400-000, Sergipe, Brazil. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institutional Program of Scientific Dissemination–IFES (PRODIF-IFES), FAPITEC/SE/FUNTEC.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The accurate assessment of plant disease severity is crucial for effective crop management. Deep learning, especially via CNNs, is widely used for image segmentation in plant lesion detection, but accurately assessing disease severity across varied environmental conditions remains challenging. This study evaluates eight deep learning models for detecting and quantifying Cercospora leaf spot (Cercospora capsici) severity in chili peppers under natural field conditions. A custom dataset of 1645 chili pepper leaf images, collected from a Brazilian plantation and annotated with 6282 lesions, was developed for real-world robustness, reflecting real-world variability in lighting and background. First, an algorithm was developed to process raw images, applying ROI selection and background removal. Then, four YOLOv8 and four Mask R-CNN models were fine-tuned for pixel-level segmentation and severity classification, comparing one-stage and two-stage models to offer practical insights for agricultural applications. In pixel-level segmentation on the test dataset, Mask R-CNN achieved superior precision with a Mean Intersection over Union (MIoU) of 0.860 and F1-score of 0.924 for the mask_rcnn_R101_FPN_3x model, compared to 0.808 and 0.893 for the YOLOv8s-Seg model. However, in severity classification, Mask R-CNN underestimated higher severity levels, with an accuracy of 72.3% for level III, while YOLOv8 attained 91.4%. Additionally, YOLOv8 demonstrated greater efficiency, with an inference time of 27 ms versus 89 ms for Mask R-CNN. While Mask R-CNN excels in segmentation accuracy, YOLOv8 offers a compelling balance of speed and reliable severity classification, making it suitable for real-time plant disease assessment in agricultural applications.

Schlagwörter

Stichwörter der Autor:innen

CNNcercospora leaf spotconvolutional neural networksplant disease severity

Autor:innen und Institutionen

  • EMEC, Sergipe Educational, Technology and Scientific Institute, Lourival Batista, Lourival Batista Highway s/n, Lagarto 49400-000, Sergipe, Brazil.
  • Laboratory of Embedded Systems and Robotics, Paraíba Federal University, Campus I Lot. Cidade Universitaria, João Pessoa 58051-900, Paraıba, Brazil.
  • Development and Environment Post-Graduation Program, Sergipe Federal University, Marcelo Déda Chagas Avenue, São Cristóvão 49107-230, Sergipe, Brazil.
  • Postgraduate Program in Agroecology, Federal Institute of Espírito Santo (IFES), Santa Teresa Campus, ES-080 Highway, Km 93, São João de Petrópolis, Santa Teresa 29660-000, Espírito Santo, Brazil.

ORCID-Profile: Douglas Vieira Leite, Alisson Vasconcelos de Brito, Gregorio Guirada Faccioli, Gustavo Haddad Souza Vieira

Förderung

  • Institutional Program of Scientific Dissemination–IFES (PRODIF-IFES)
  • FAPITEC/SE/FUNTEC

Zitieren

APA

Leite, D. V., de Brito, A. V., Faccioli, G. G., & Haddad Souza Vieira, G. (2025). Deep Learning Models for Detection and Severity Assessment of Cercospora Leaf Spot (Cercospora capsici) in Chili Peppers Under Natural Conditions. Plants, 14(13), 2011. https://doi.org/10.3390/plants14132011

Vancouver

Leite DV, de Brito AV, Faccioli GG, Haddad Souza Vieira G. Deep Learning Models for Detection and Severity Assessment of Cercospora Leaf Spot (Cercospora capsici) in Chili Peppers Under Natural Conditions. Plants. 2025;14(13):2011. doi:10.3390/plants14132011

BibTeX
@article{Leite2025,
  author = {Leite, Douglas Vieira and de Brito, Alisson Vasconcelos and Faccioli, Gregorio Guirada and Haddad Souza Vieira, Gustavo},
  title = {Deep Learning Models for Detection and Severity Assessment of Cercospora Leaf Spot (Cercospora capsici) in Chili Peppers Under Natural Conditions},
  journal = {Plants},
  year = {2025},
  volume = {14},
  number = {13},
  pages = {2011},
  doi = {10.3390/plants14132011}
}
Kostenlos starten →