EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

VLM-MCPDD: An Interpretable Vision Language Model for Multi-Crop Pests and Disease Diagnosis

Liang Zhao, Mengwei Li, Xu Ren, Yuting Cheng, Zongxi Hu

Applied Sciences · 05.06.2026 · 16 (11) · 5719

DOI: 10.3390/app16115719 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 05.06.2026. Verfasst von Liang Zhao und 4 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Natural Science Foundation of China, the Key Scientific Research Projects of Higher Education Institutions in Henan Province, the Open Fund of Key Laboratory of Grain Information Processing and Control (Henan University of Technology), Ministry of Education.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep convolutional neural networks have made substantial progress in automated crop disease diagnosis. However, their practical application remains constrained by limited interpretability and insufficient structured reasoning, as these models largely operate as black boxes. Although they are effective in extracting visual features, they often fail to provide semantically grounded explanations, which may reduce their reliability in complex and open agricultural environments. To address these issues, this study constructs a Vision Language Model for Multi-Crop Pest and Disease Diagnosis (VLM-MCPDD). Specifically, the LLaVA-1.5 model is fine-tuned using low-rank adaptation (LoRA) to better align visual symptom representations with domain-specific agricultural knowledge. In addition, a Pests and Diseases Semantic Dataset (PDSD) is constructed to support multimodal learning. Based on PDSD, a chain-of-thought (CoT) mechanism is introduced to simulate the diagnostic workflow of agronomists, covering symptom observation, causal analysis, and final decision-making. The experimental results show that compared with comparative models such as Swin Transformer and ConvNeXt, VLM-MCPDD performs better in overall performance and can provide some reference for disease and pest diagnosis in intelligent agriculture.

Förderung

  • the National Natural Science Foundation of China
  • the Key Scientific Research Projects of Higher Education Institutions in Henan Province
  • the Open Fund of Key Laboratory of Grain Information Processing and Control (Henan University of Technology), Ministry of Education

Zitieren

APA

Zhao, L., Li, M., Ren, X., Cheng, Y., & Hu, Z. (2026). VLM-MCPDD: An Interpretable Vision Language Model for Multi-Crop Pests and Disease Diagnosis. Applied Sciences, 16(11), 5719. https://doi.org/10.3390/app16115719

Vancouver

Zhao L, Li M, Ren X, Cheng Y, Hu Z. VLM-MCPDD: An Interpretable Vision Language Model for Multi-Crop Pests and Disease Diagnosis. Applied Sciences. 2026;16(11):5719. doi:10.3390/app16115719

BibTeX
@article{Zhao2026,
  author = {Zhao, Liang and Li, Mengwei and Ren, Xu and Cheng, Yuting and Hu, Zongxi},
  title = {VLM-MCPDD: An Interpretable Vision Language Model for Multi-Crop Pests and Disease Diagnosis},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {11},
  pages = {5719},
  doi = {10.3390/app16115719}
}
Kostenlos starten →