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 Video

Updated: Jul 16, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Multi-Spectral Band Analysis for Satellite-to-Aerial Image Registration: A Comparative Study of Deep Learning and

Dongyeob Han1, Jeong Heon Song2, Sun-Gu Lee2

  • 1Department of Civil Engineering, Chonnam National University, Gwangju 61186, Republic of Korea.

Sensors (Basel, Switzerland)
|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

Rectification of Bowl-Shape Deformation of Tidal Flat DEM derived from UAV Imaging.

Sensors (Basel, Switzerland)ยท2020
See all related articles

This study evaluated multi-spectral band performance for image registration between satellite and aerial imagery for autonomous vehicle navigation. Detector-free methods offered stable correspondences, but independent accuracy was similar across all tested matchers.

Area of Science:

  • Geospatial Information Science
  • Computer Vision
  • Remote Sensing

Background:

  • High-definition (HD) map generation for autonomous vehicles requires precise geometric registration between satellite imagery and aerial orthophotos.
  • Evaluating multi-spectral band performance is crucial for optimizing this registration process.

Purpose of the Study:

  • To comprehensively evaluate multi-spectral band performance for image registration between KOMPSAT-3A satellite imagery and VWorld aerial orthophotos.
  • To compare various feature-matching approaches, including LightGlue, SIFT, edge-based FFT, LoFTR, and RoMa, across different patch sizes.

Main Methods:

  • Systematic comparison of five feature-matching approaches: LightGlue, edge-based FFT, and SIFT-based methods, all with and without CLAHE preprocessing.
  • Integration and comparison of two detector-free deep matchers, LoFTR and RoMa.
Keywords:
CLAHEHD mapKOMPSAT-3ALightGlueLoFTRRoMaSIFTfeature matchingimage registrationmulti-spectral analysis

Related Experiment Videos

Last Updated: Jul 16, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

  • Evaluation across seven patch size configurations and multiple spectral bands.

    • Main Results:

    • Panchromatic-derived products (SPECPAN, EMPPAN) and BT601 luminance composite showed superior registration stability.
    • LightGlue achieved high inlier counts (1100+), while SIFT with CLAHE had the lowest RMSE (1.55 pixels).
    • Detector-free methods (LoFTR, RoMa) yielded dense, stable correspondences; LoFTR demonstrated the best transformation stability.

    Conclusions:

    • Matcher selection impacts correspondence density and transformation stability, not independent geodetic accuracy (approx. 2.8 m).
    • Achieved meter-level accuracy is suitable for HD map preprocessing and Ground Control Point (GCP) generation.
    • Guidance is specific to the KOMPSAT-3A/VWorld dataset and registration task.