EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DEMNet: A Small Object Detection Method for Tea Leaf Blight in Slightly Blurry UAV Remote Sensing Images

Yating Gu, Yuxin Jing, Hao-Dong Li, Juntao Shi, Haifeng Lin

Remote Sensing · 06.06.2025 · 17 (12) · 1967

DOI: 10.3390/rs17121967 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 06.06.2025. Verfasst von Yating Gu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Qinglan Project 2024 and the key core technology industrialization project of the whole industrial chain of agricultural characteristic industries in Nanjing.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unmanned aerial vehicles are widely used in agricultural disease detection. Still, slight image blurring caused by lighting, wind, and flight instability often hampers the detection of dense small targets like tea leaf blight spots. In response to this problem, this paper proposes DEMNet, a model based on the YOLOv8n architecture. The goal is to enhance small, blurry object detection performance in UAV-based scenarios. DEMNet introduces a dynamic convolution mechanism into the HGNetV2 backbone to form DynamicHGNetV2, enabling adaptive convolutional weight generation and improving feature extraction for blurry objects. An efficient EMAFPN neck structure further facilitates deep–shallow feature interaction while reducing the computational cost. Additionally, a novel CMLAB module replaces the traditional C2f structure, employing multi-scale convolutions and local attention mechanisms to recover semantic information in blurry regions and better detect densely distributed small targets. Experimental results on a slightly blurry tea leaf blight dataset demonstrate that DEMNet surpasses the baseline by 5.7% in recall and 4.9% in mAP@0.5. Moreover, the model reduces parameters to 1.7 M, computation to 6.1 GFLOPs, and model size to 4.2 MB, demonstrating high accuracy and strong deployment potential.

Förderung

  • Qinglan Project 2024 and the key core technology industrialization project of the whole industrial chain of agricultural characteristic industries in Nanjing

Zitieren

APA

Gu, Y., Jing, Y., Li, H., Shi, J., & Lin, H. (2025). DEMNet: A Small Object Detection Method for Tea Leaf Blight in Slightly Blurry UAV Remote Sensing Images. Remote Sensing, 17(12), 1967. https://doi.org/10.3390/rs17121967

Vancouver

Gu Y, Jing Y, Li H, Shi J, Lin H. DEMNet: A Small Object Detection Method for Tea Leaf Blight in Slightly Blurry UAV Remote Sensing Images. Remote Sensing. 2025;17(12):1967. doi:10.3390/rs17121967

BibTeX
@article{Gu2025,
  author = {Gu, Yating and Jing, Yuxin and Li, Hao-Dong and Shi, Juntao and Lin, Haifeng},
  title = {DEMNet: A Small Object Detection Method for Tea Leaf Blight in Slightly Blurry UAV Remote Sensing Images},
  journal = {Remote Sensing},
  year = {2025},
  volume = {17},
  number = {12},
  pages = {1967},
  doi = {10.3390/rs17121967}
}
Kostenlos starten →