EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SwinPyramidNet: a deep learning network for pear leaf disease classification using hybrid convolutional feature pyramids and swin transformer

Tejas Sharma, Nikhil Shinde, J. Mukundh, R. Karthik, K. Suganthi

Frontiers in Signal Processing · 19.06.2026 · 6

DOI: 10.3389/frsip.2026.1821900 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Signal Processing, online erschienen am 19.06.2026. Verfasst von Tejas Sharma und 4 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Pear is rich in dietary fiber, vitamins, and minerals, but its cultivation has been affected by the adverse effects of climate change. Leaf health serves as an early indicator of potential fruit damage, making timely disease detection essential to prevent substantial economic losses. Conventional diagnosis methods, relying on manual visual inspection, are often labor-intensive, costly, and prone to errors. Recent advances in deep learning have enabled accurate image-based disease classification, providing an effective alternative for automated disease detection in precision agriculture.

Methods

This study introduces SwinPyramidNet , a novel dual-track deep learning architecture for pear leaf disease classification. The network integrates the strengths of Convolutional Neural Networks (CNNs) and Transformers to enhance classification performance. The Swin Transformer branch captures global contextual information by modeling long-range dependencies and complex patterns, while the Hybrid Convolutional Feature Pyramid Network (HCFPNet) branch focuses on extracting fine-grained local features through multi-scale convolutional processing. The outputs from both branches are fused and refined using a coordinate attention mechanism, which encodes positional information to emphasize spatially relevant features across channels. This complementary feature extraction strategy enables the network to effectively learn both global and local disease characteristics from pear leaf images.

Results

The proposed network achieved a classification accuracy of 92.69% on the Diamos dataset. The experimental findings demonstrate the effectiveness of combining Transformer-based global feature learning with CNN-based local feature extraction for pear leaf disease classification. The coordinate attention module further contributed to improved feature representation by highlighting disease-relevant regions.

Discussion

The obtained results indicate that SwinPyramidNet can serve as a reliable and efficient approach for automated pear leaf disease diagnosis. The integration of global contextual information and fine-grained local features enhances the network’s ability to distinguish between disease classes with high accuracy. These findings highlight the applicability of hybrid CNN–Transformer architectures in smart agriculture and plant disease management systems.

Zitieren

APA

Sharma, T., Shinde, N., Mukundh, J., Karthik, R., & Suganthi, K. (2026). SwinPyramidNet: a deep learning network for pear leaf disease classification using hybrid convolutional feature pyramids and swin transformer. Frontiers in Signal Processing, 6. https://doi.org/10.3389/frsip.2026.1821900

Vancouver

Sharma T, Shinde N, Mukundh J, Karthik R, Suganthi K. SwinPyramidNet: a deep learning network for pear leaf disease classification using hybrid convolutional feature pyramids and swin transformer. Frontiers in Signal Processing. 2026;6. doi:10.3389/frsip.2026.1821900

BibTeX
@article{Sharma2026,
  author = {Sharma, Tejas and Shinde, Nikhil and Mukundh, J. and Karthik, R. and Suganthi, K.},
  title = {SwinPyramidNet: a deep learning network for pear leaf disease classification using hybrid convolutional feature pyramids and swin transformer},
  journal = {Frontiers in Signal Processing},
  year = {2026},
  volume = {6},
  doi = {10.3389/frsip.2026.1821900}
}
Kostenlos starten →