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: Oct 29, 2025

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
07:13

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

4.0K

Predicting road quality using high resolution satellite imagery: A transfer learning approach.

Ethan Brewer1, Jason Lin1, Peter Kemper2

  • 1Department of Applied Science, William & Mary, Williamsburg, VA, United States of America.

Plos One
|July 9, 2021
PubMed
Summary

Related Concept Videos

Design Example: Alignment of a Road Line Using GIS01:17

Design Example: Alignment of a Road Line Using GIS

147
The alignment of a road line using Geographic Information Systems (GIS) is a critical process in civil engineering, combining advanced technology with practical decision-making. This methodology begins with the collection of geospatial data, including information on land cover, geomorphology, drainage patterns, slope, and contour details. Such data is typically acquired through satellite imagery and GIS tools, offering a comprehensive understanding of the terrain.Once the data is gathered, it...
147

You might also read

Related Articles

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

Sort by
Same author

Expandable corpectomy cage with articulating endcap for correction of severe cervical kyphotic deformity: a case report.

Journal of spine surgery (Hong Kong)·2026
Same author

PhaseOne Hypochlorous Acid is Effective for Preventing Capsular Contracture and Infection in Primary Breast Augmentation.

Aesthetic surgery journal·2026
Same author

The Ethical Landscape of Plastic Surgeons Prescribing Weight Loss Medications.

Aesthetic surgery journal·2026
Same author

Bioinspired Nanofibers Incorporating Radiocatalytic Nanoparticles to Capture Migratory Glioma Cells for Localized Radiosensitization and Inhibition of Tumor Recurrence.

Advanced healthcare materials·2026
Same author

Tumor cell death by ferroptosis contributes to an immunosuppressive tumor microenvironment in syngeneic murine models of cancer.

Cancer & metabolism·2026
Same author

Genomic and Transcriptomic Analysis of High-Grade Endometrial Carcinoma Reveals Biological Heterogeneity and Molecular Classification Challenges.

Cancer research communications·2026

This study uses satellite imagery and artificial intelligence to assess road quality and travel speed, improving infrastructure investment decisions globally. The developed models achieved high accuracy in the US and Nigeria.

Area of Science:

  • Geospatial analysis
  • Machine learning applications in infrastructure assessment
  • Transportation engineering

Background:

  • Road infrastructure is crucial for human health and economic development, yet accurate data on existing roads is often lacking.
  • Previous research utilized satellite imagery for road detection and mapping.
  • There is a need for methods to assess road quality and travel speed using remote sensing data.

Purpose of the Study:

  • To extend existing literature by using satellite imagery to estimate road quality and travel speed.
  • To develop and evaluate transfer learning approaches using convolutional neural networks (CNNs) for road quality assessment.
  • To assess the effectiveness of these models in diverse geographical contexts, specifically the US and Nigeria.

Main Methods:

Related Experiment Videos

Last Updated: Oct 29, 2025

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
07:13

Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy

Published on: February 25, 2021

4.0K
  • A transfer learning approach was employed, pre-training CNN architectures on US road data and fine-tuning on Nigerian data.
  • Eight different CNN architectures were tested using a dataset of 53,686 images covering 2,400 km of US roads, classified by quality (low, middle, high).
  • The best-performing model was adapted for a case study in Nigeria using 1,000 images of paved and unpaved roads.

Main Results:

  • Satellite imagery estimation of road quality in the US achieved 80.0% accuracy, with 99.4% of predictions within one class.
  • Fine-tuning the US-trained model on Nigerian data resulted in 94.0% accuracy for predicting road quality.
  • The model demonstrated an average prediction accuracy within 0.32 on a 0-3 scale for road quality.

Conclusions:

  • Satellite imagery combined with transfer learning offers a viable method for estimating road quality and travel speed.
  • This approach can significantly improve the identification of areas needing infrastructure investment, particularly in data-scarce regions.
  • The transfer learning methodology shows promise for global application in road infrastructure assessment and planning.