EUVIMEDEUVIMED entdecken
Open Access

Characteristic wavelength selection for rice blast based on hyperspectral remote sensing and deep convolutional neural networks

Yashi Wang, Hongze Zhang, Shuaipeng Wang, Yongsheng Liu, Jiayi Li, Shuai Feng, Chunling Chen

Pest Management Science · 2026

DOI: 10.1002/ps.70994 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Pest Management Science, 2026. Verfasst von Yashi Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang, China. Hauptthemen laut MEDLINE: Oryza, Remote Sensing Technology, Plant Diseases, Hyperspectral Imaging. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: [Liaoning Provincial Department of Education Basic Scientific Research Project-General Project (Science)] grant number [LJ212510157019].

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND Hyperspectral remote sensing technology is one of the key technical methods for detecting rice blast in the field, but existing hyperspectral dimensionality reduction methods still suffer from information redundancy and insufficient feature interpretability. This study aimed to develop a feature wavelength selection method integrating deep learning and model attribution analysis to extract key spectral features across different disease severity levels. RESULTS A residual network model Dilated Convolution and Deformable Convolution‐Residual Network (DCR‐ResNet) combining dilated convolution and deformable convolution was constructed to deeply mine spectral features across varying disease severities. Meanwhile, the Integrated Gradient (IG) and Gradient‐weighted Class Activation Mapping (Grad‐CAM) methods were combined to enable the selection of spectral wavelengths. The effectiveness of the proposed method was validated using statistical analysis (transformed divergence, within‐class scatter) and modeling analysis. Findings reveal that the spectral feature wavelengths identified by DCR‐ResNet in conjunction with the IG‐GradCAM approach exhibit excellent inter‐class separability and intra‐class compactness. Furthermore, when benchmarked against conventional dimensionality reduction techniques such as Successive Projections Algorithm, Random Frog, and Competitive Adaptive Reweighted Sampling, the Support Vector Machine, Extreme Learning Machine, and Random Forest models developed using IG‐GradCAM‐selected feature wavelengths demonstrate superior classification performance. The overall accuracy reaches 85.9%, 85.5% and 86.2%, with kappa values of 81.3%, 80.6% and 81.6%, respectively. CONCLUSION The feature wavelength selection method combining DCR‐ResNet with IG‐GradCAM not only improves the accuracy of hyperspectral feature extraction but also provides an efficient and feasible approach for the precise identification of rice blast. © 2026 Society of Chemical Industry.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OryzaRemote Sensing TechnologyPlant DiseasesHyperspectral ImagingConvolutional Neural NetworksDeep Learning

Stichwörter der Autor:innen

characteristic wavelengthdeep learninghyperspectralmodel attribution analysisrice blast

Autor:innen und Institutionen

  • School of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang, China.
  • Natl Digital Agr Reg Innovat Sub Ctr Northeast, Shenyang, China.
  • Liaoning Provincial Key Laboratory of Smart Agricultural Technology, Shenyang, China.

Förderung

  • [Liaoning Provincial Department of Education Basic Scientific Research Project-General Project (Science)] grant number [LJ212510157019]

Zitieren

APA

Wang, Y., Zhang, H., Wang, S., Liu, Y., Li, J., Feng, S., & Chen, C. (2026). Characteristic wavelength selection for rice blast based on hyperspectral remote sensing and deep convolutional neural networks. Pest Management Science. https://doi.org/10.1002/ps.70994

Vancouver

Wang Y, Zhang H, Wang S, Liu Y, Li J, Feng S, et al. Characteristic wavelength selection for rice blast based on hyperspectral remote sensing and deep convolutional neural networks. Pest Management Science. 2026. doi:10.1002/ps.70994

BibTeX
@article{Wang2026,
  author = {Wang, Yashi and Zhang, Hongze and Wang, Shuaipeng and Liu, Yongsheng and Li, Jiayi and Feng, Shuai and Chen, Chunling},
  title = {Characteristic wavelength selection for rice blast based on hyperspectral remote sensing and deep convolutional neural networks},
  journal = {Pest Management Science},
  year = {2026},
  doi = {10.1002/ps.70994}
}
Kostenlos starten →