EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Supervised Hyperspectral Band Selection Using Texture Features for Classification of Citrus Leaf Diseases with YOLOv8

Quentin Frederick iD, Thomas Burks, Jonathan Adam Watson iD, Pappu Kumar Yadav iD, Jianwei Qin iD, Moon Kim, Megan M Dewdney iD

Sensors · 09.02.2025 · 25 (4) · 1034

DOI: 10.3390/s25041034 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 09.02.2025. Verfasst von Quentin Frederick und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Agricultural and Biological Engineering, University of Florida, P.O. Box 110570, Gainesville, FL 32611-0570, USA. Hauptthemen laut MEDLINE: Citrus, Plant Diseases, Plant Leaves, Hyperspectral Imaging. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: USDA ARS Cooperative Agreement.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Citrus greening disease (HLB) and citrus canker cause financial losses in Florida citrus groves via smaller fruits, blemishes, premature fruit drop, and/or eventual tree death. Management of these two diseases requires early detection and distinction from other leaf defects and infections. Automated leaf inspection with hyperspectral imagery (HSI) is tested in this study. Citrus leaves bearing visible symptoms of HLB, canker, scab, melanose, greasy spot, zinc deficiency, and a control class were collected, and images were taken with a line-scan HSI camera. YOLOv8 was trained to classify multispectral images from this image dataset, created by selecting bands with a novel variance-based method. The ‘small’ network using an intensity-based band combination yielded an overall weighted F1 score of 0.8959, classifying HLB and canker with F1 scores of 0.788 and 0.941, respectively. The network size appeared to exert greater influence on performance than the HSI bands selected. These findings suggest that YOLOv8 relies more heavily on intensity differences than on the texture properties of citrus leaves and is less sensitive to the choice of wavelengths than traditional machine vision classifiers.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CitrusPlant DiseasesPlant LeavesHyperspectral ImagingImage Processing, Computer-Assisted

Stichwörter der Autor:innen

YOLOv8band selectioncitrus cankercitrus inspectionfeature extractionhuanglongbinghyperspectral imagery (HSI)leaf inspection

Autor:innen und Institutionen

  • Department of Agricultural and Biological Engineering, University of Florida, P.O. Box 110570, Gainesville, FL 32611-0570, USA.
  • Department of Agricultural and Biosystems Engineering, South Dakota State University, P.O. Box 2100, Brookings, SD 57007, USA.
  • USDA/ARS Environmental Microbial and Food Safety Laboratory (EMFSL), Beltsville Agricultural Research Center, Beltsville, MD 20705, USA.
  • Citrus Research and Education Center, 700 Experiment Station Rd., Lake Alfred, FL 33850, USA.

ORCID-Profile: Quentin Frederick, Jonathan Adam Watson, Pappu Kumar Yadav, Jianwei Qin, Megan M Dewdney

Förderung

  • USDA ARS Cooperative Agreement

Zitieren

APA

Frederick, Q., Burks, T., Watson, J. A., Yadav, P. K., Qin, J., Kim, M., & Dewdney, M. M. (2025). Supervised Hyperspectral Band Selection Using Texture Features for Classification of Citrus Leaf Diseases with YOLOv8. Sensors, 25(4), 1034. https://doi.org/10.3390/s25041034

Vancouver

Frederick Q, Burks T, Watson JA, Yadav PK, Qin J, Kim M, et al. Supervised Hyperspectral Band Selection Using Texture Features for Classification of Citrus Leaf Diseases with YOLOv8. Sensors. 2025;25(4):1034. doi:10.3390/s25041034

BibTeX
@article{Frederick2025,
  author = {Frederick, Quentin and Burks, Thomas and Watson, Jonathan Adam and Yadav, Pappu Kumar and Qin, Jianwei and Kim, Moon and Dewdney, Megan M},
  title = {Supervised Hyperspectral Band Selection Using Texture Features for Classification of Citrus Leaf Diseases with YOLOv8},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {4},
  pages = {1034},
  doi = {10.3390/s25041034}
}
Kostenlos starten →