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

Superpixel Random Selection Random Walk Multi-Branch Depthwise Convolutional Neural Network for Hyperspectral Image

Kai Zhang1, Xinwei Jiang1, Zhihua Cai1

  • 1School of Computer Science, China University of Geosciences, Wuhan 430074, China.

Sensors (Basel, Switzerland)
|June 12, 2026
PubMed
Summary
This summary is machine-generated.

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

Sparse Feature Learning of Hyperspectral Imagery via Multiobjective-Based Extreme Learning Machine.

Sensors (Basel, Switzerland)·2020
Same author

Spectral-Spatial Feature Extraction of Hyperspectral Images Based on Propagation Filter.

Sensors (Basel, Switzerland)·2018
Same author

Triptolide protects rat heart against pressure overload-induced cardiac fibrosis.

International journal of cardiology·2013
Same author

Multiresidue pesticide analysis of botanical dietary supplements using salt-out acetonitrile extraction, solid-phase extraction cleanup column, and gas chromatography-triple quadrupole mass spectrometry.

Analytical chemistry·2013
Same author

Interaction domains of p62: a bridge between p62 and selective autophagy.

DNA and cell biology·2013
Same author

Predictors of seizure freedom after surgical management of tuberous sclerosis complex: a systematic review and meta-analysis.

Epilepsy research·2013
Same journal

RETRACTED: Zhang et al. A Novel Framework for Reconstruction and Imaging of Target Scattering Centers via Wide-Angle Incidence in Radar Networks. <i>Sensors</i> 2025, <i>25</i>, 6802.

Sensors (Basel, Switzerland)·2026
Same journal

Enhancing Unsupervised Multi-Source Domain Adaptation for Person Re-Identification via Mixture of Experts and Graph-Based Relation.

Sensors (Basel, Switzerland)·2026
Same journal

Development of an Instrumented Glove for Palmar Pressure Assessment in Kayakers.

Sensors (Basel, Switzerland)·2026
Same journal

Development and Experimental Validation of an Autonomous IoT-Based Monitoring System for Real-Time Water Quality Assessment in the Amazon River.

Sensors (Basel, Switzerland)·2026
Same journal

Semi-Supervised Adversarial Learning Framework for Controller Area Network Bus Intrusion Detection.

Sensors (Basel, Switzerland)·2026
Same journal

Smart Optimization Method for Safety Signs in Innovative Manufacturing Environments Integrating Industrial Field IoT Sensors and Knowledge Graphs.

Sensors (Basel, Switzerland)·2026
See all related articles

This study introduces a novel training-free Convolutional Neural Network (CNN) for hyperspectral image (HSI) processing. The proposed SRSRWMD-CNN method enhances feature extraction, improving HSI classification accuracy without extensive training.

Area of Science:

  • Remote Sensing
  • Computer Vision
  • Artificial Intelligence

Background:

  • Convolutional Neural Networks (CNNs) are effective for hyperspectral image (HSI) processing.
  • Training-free CNN variants offer reduced parameterization but suffer from noise and redundancy in kernels.
  • Existing methods struggle with efficient and accurate feature extraction in HSI analysis.

Purpose of the Study:

  • To propose a novel training-free CNN for HSI processing that overcomes limitations of existing methods.
  • To enhance feature extraction capabilities for improved HSI classification.
  • To develop a method that reduces computational cost by avoiding extensive model training.

Main Methods:

  • Introduced Superpixel Random Selection Random Walk Multi-Branch Depthwise Convolutional Neural Network (SRSRWMD-CNN).
Keywords:
CNNHSIconvolutional neural networkdeep learninghyperspectral image classification

Related Experiment Videos

  • Employed multi-scale superpixel segmentation to generate training-free convolution kernels.
  • Utilized a multi-branch depthwise convolution strategy and a random walk strategy for feature enhancement and robustness.
  • Integrated multi-scale spectral-spatial features across multiple convolutional stages.

    • Main Results:

    • The SRSRWMD-CNN method achieved superior performance in HSI classification compared to state-of-the-art algorithms.
    • Demonstrated effective multi-scale spectral-spatial feature representation without costly training.
    • Showcased enhanced feature learning and robustness through proposed strategies.

    Conclusions:

    • The proposed SRSRWMD-CNN is a highly effective training-free approach for HSI classification.
    • The method successfully addresses noise and redundancy issues in training-free kernels.
    • SRSRWMD-CNN offers a promising direction for efficient and accurate hyperspectral image analysis.