Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Defective wheat kernel classification using dual-range hyperspectral imaging and an interpretable spectral-spatial

Dianyang Sun1, Weijie Lan1, Kang Tu1

  • 1College of Food Science and Technology, Nanjing Agricultural University, Nanjing, Jiangsu, 210095, China.

Spectrochimica Acta. Part A, Molecular and Biomolecular Spectroscopy
|July 15, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Non-destructive prediction of the moisture content of individual wheat kernels combining hyperspectral imaging and WGAN data augmentation algorithm.

Food research international (Ottawa, Ont.)·2025
Same author

Discrimination of unsound soybeans using hyperspectral imaging: A deep learning method based on dual-channel feature fusion strategy and attention mechanism.

Food research international (Ottawa, Ont.)·2025
Same author

In vivo tracking of superparamagnetic iron oxide nanoparticle-labeled mesenchymal stem cell tropism to malignant gliomas using magnetic resonance imaging. Laboratory investigation.

Journal of neurosurgery·2008
Same author

Enhancement and broadening of extreme-ultraviolet supercontinuum in a relative phase controlled two-color laser field.

Optics letters·2008
Same author

Screening and breeding of high taxol producing fungi by genome shuffling.

Science in China. Series C, Life sciences·2008
Same author

Reversible self-association of a concentrated monoclonal antibody solution mediated by Fab-Fab interaction that impacts solution viscosity.

Journal of pharmaceutical sciences·2008

A new spectral-spatial fusion convolutional neural network (SSFCNN) accurately identifies defective wheat kernels using hyperspectral imaging. This advanced method improves quality control for wheat screening.

Area of Science:

  • Agricultural Science
  • Computer Vision
  • Spectroscopy

Background:

  • Accurate detection of defective wheat kernels is crucial for quality control.
  • Existing methods struggle with subtle spectral and spatial differences, limiting reliability.
  • Hyperspectral imaging offers rich spectral and spatial data for kernel analysis.

Purpose of the Study:

  • To develop a novel spectral-spatial fusion convolutional neural network (SSFCNN) for classifying defective wheat kernels.
  • To integrate complementary spectral and spatial information for enhanced classification accuracy.
  • To provide a validated framework for intelligent wheat grading and online quality control.

Main Methods:

  • A spectral-spatial fusion convolutional neural network (SSFCNN) was designed for wheat kernel classification.
Keywords:
Attention mechanismDefective wheat kernelsDual-range hyperspectral imagingModel interpretabilitySpectral-spatial fusion

Related Experiment Videos

  • Attention mechanisms (SE, SA, ECA) were integrated for spectral channel recalibration and feature enhancement.
  • The SSFCNN performance was compared against Support Vector Machine (SVM) and conventional Convolutional Neural Network (CNN) methods.

    • Main Results:

    • The SSFCNN achieved high accuracies: 96.48% (Vis-NIR) and 95.61% (SWIR).
    • The SSFCNN significantly outperformed SVM and conventional CNN approaches in precision, recall, specificity, and F1-score.
    • Visualizations confirmed improved spatial coherence and decision reliability of the SSFCNN model.

    Conclusions:

    • The developed SSFCNN provides a validated and interpretable method for hyperspectral screening of defective wheat kernels.
    • This approach offers a strong methodological basis for future intelligent grading and online quality control systems.
    • Further validation under real-world sorting conditions is recommended for practical implementation.