EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

I-GhostNetV3: A Lightweight Deep Learning Framework for Vision-Sensor-Based Rice Leaf Disease Detection in Smart Agriculture

Puyu Zhang iD, Rui Li, Yuxuan Liu, Guoxi Sun, Chenglin Wen

Sensors · 04.02.2026 · 26 (3) · 1025

DOI: 10.3390/s26031025 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 04.02.2026. Verfasst von Puyu Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Maoming 525000, China. Hauptthemen laut MEDLINE: Oryza, Deep Learning, Plant Diseases, Plant Leaves. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Guangdong Provincial Key Area Research and Development Program, the National Key Research and Development Program, the Sci-Tech Innovation Project of the Graduate School of Guangdong University of Petrochemical Technolog.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate and timely diagnosis of rice leaf diseases is crucial for smart agriculture leveraging vision sensors. However, existing lightweight convolutional neural networks (CNNs) often struggle in complex field environments, where small lesions, cluttered backgrounds, and varying illumination complicate recognition. This paper presents I-GhostNetV3, an incrementally improved GhostNetV3-based network for RGB rice leaf disease recognition. I-GhostNetV3 introduces two modular enhancements with controlled overhead: (1) Adaptive Parallel Attention (APA), which integrates edge-guided spatial and channel cues and is selectively inserted to enhance lesion-related representations (at the cost of additional computation), and (2) Fusion Coordinate-Channel Attention (FCCA), a near-neutral SE replacement that enables efficient spatial–channel feature fusion to suppress background interference. Experiments on the Rice Leaf Bacterial and Fungal Disease (RLBF) dataset show that I-GhostNetV3 achieves 90.02% Top-1 accuracy with 1.831 million parameters and 248.694 million FLOPs, outperforming MobileNetV2 and EfficientNet-B0 under our experimental setup while remaining compact relative to the original GhostNetV3. In addition, evaluation on PlantVillage-Corn serves as a supplementary transfer sanity check; further validation on independent real-field target domains and on-device profiling will be explored in future work. These results indicate that I-GhostNetV3 is a promising efficient backbone for future edge deployment in precision agriculture.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OryzaDeep LearningPlant DiseasesPlant LeavesAgricultureConvolutional Neural Networks

Stichwörter der Autor:innen

I-GhostNetV3attention mechanismslightweight convolutional networksrice disease classificationsmart agriculturevision sensors

Autor:innen und Institutionen

  • College of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Maoming 525000, China.
  • Guangdong Provincial Key Laboratory of Petrochemical Equipment Fault Diagnosis, Guangdong University of Petrochemical Technology, Maoming 525000, China.

ORCID-Profile: Puyu Zhang

Förderung

  • the Guangdong Provincial Key Area Research and Development Program
  • the National Key Research and Development Program
  • the Sci-Tech Innovation Project of the Graduate School of Guangdong University of Petrochemical Technolog
  • Guangdong Provincial Key Area Research and Development Program
  • National Key Research and Development Program
  • Sci-Tech Innovation Project of the Graduate School of Guangdong University of Petrochemical Technolog

Zitieren

APA

Zhang, P., Li, R., Liu, Y., Sun, G., & Wen, C. (2026). I-GhostNetV3: A Lightweight Deep Learning Framework for Vision-Sensor-Based Rice Leaf Disease Detection in Smart Agriculture. Sensors, 26(3), 1025. https://doi.org/10.3390/s26031025

Vancouver

Zhang P, Li R, Liu Y, Sun G, Wen C. I-GhostNetV3: A Lightweight Deep Learning Framework for Vision-Sensor-Based Rice Leaf Disease Detection in Smart Agriculture. Sensors. 2026;26(3):1025. doi:10.3390/s26031025

BibTeX
@article{Zhang2026,
  author = {Zhang, Puyu and Li, Rui and Liu, Yuxuan and Sun, Guoxi and Wen, Chenglin},
  title = {I-GhostNetV3: A Lightweight Deep Learning Framework for Vision-Sensor-Based Rice Leaf Disease Detection in Smart Agriculture},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {3},
  pages = {1025},
  doi = {10.3390/s26031025}
}
Kostenlos starten →