EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Lightweight Edge-Deployable Framework for Intelligent Rice Disease Monitoring Based on Pruning and Distillation

Wei Liu, Baoquan Duan, Zhipeng Fan, Ming Chen, Zeguo Qiu

Sensors · 2025

DOI: 10.3390/s26010035 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Wei Liu und 4 weiteren Autor:innen. Beteiligte Einrichtung: Harbin University of Commerce, Harbin 150028, China. Hauptthemen laut MEDLINE: Oryza, Plant Diseases. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Heilongjiang Province, Heilongjiang Postdoctoral Fund to pursue scientific research, Heilongjiang Province Colleges and Universities Basic Scientific Research Business Expenses Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Digital agriculture and smart farming require crop health monitoring methods that balance detection accuracy with computational cost. Rice leaf diseases threaten yield, while field images often contain small multi-scale lesions, variable illumination and cluttered backgrounds. This paper investigates SCD-YOLOv11n, a lightweight detector designed with these constraints in mind. The model replaces the YOLOv11n backbone with a StarNet backbone and integrates a C3k2-Star module to enhance fine-grained, multi-scale feature extraction. A Detail-Strengthened Cross-scale Detection (DSCD) head is further introduced to improve localization of small lesions. On this architecture, we design a DepGraph-based mixed group-normalization pruning rule and apply channel-wise feature distillation to recover performance after pruning. Experiments on a public rice leaf disease dataset show that the compressed model requires 1.9 MB of storage, achieves 97.4% mAP@50 and 76.2% mAP@50:95, and attains a measured speed of 184 FPS under the tested settings. These results provide a quantitative reference for designing lightweight object detectors for rice disease monitoring in digital agriculture scenarios.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OryzaPlant DiseasesAgricultureAlgorithmsDetection AlgorithmsPlant LeavesImage Processing, Computer-AssistedDistillation

Stichwörter der Autor:innen

digital agricultureknowledge distillationlightweight YOLOnetwork pruningrice leaf disease detection

Autor:innen und Institutionen

  • Harbin University of Commerce, Harbin 150028, China.
  • Heilongjiang Provincial Key Laboratory of Electronic Commerce and Information Processing, Harbin 150028, China.

Förderung

  • Natural Science Foundation of Heilongjiang Province
  • Heilongjiang Postdoctoral Fund to pursue scientific research
  • Heilongjiang Province Colleges and Universities Basic Scientific Research Business Expenses Project
  • Collaborative Innovation Achievement Program of Double First-class Disciplines in Heilongjiang Province

Zitieren

APA

Liu, W., Duan, B., Fan, Z., Chen, M., & Qiu, Z. (2025). A Lightweight Edge-Deployable Framework for Intelligent Rice Disease Monitoring Based on Pruning and Distillation. Sensors. https://doi.org/10.3390/s26010035

Vancouver

Liu W, Duan B, Fan Z, Chen M, Qiu Z. A Lightweight Edge-Deployable Framework for Intelligent Rice Disease Monitoring Based on Pruning and Distillation. Sensors. 2025. doi:10.3390/s26010035

BibTeX
@article{Liu2025,
  author = {Liu, Wei and Duan, Baoquan and Fan, Zhipeng and Chen, Ming and Qiu, Zeguo},
  title = {A Lightweight Edge-Deployable Framework for Intelligent Rice Disease Monitoring Based on Pruning and Distillation},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s26010035}
}
Kostenlos starten →