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 7, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

674

Robust Real-Time Traffic Surveillance with Deep Learning.

Jessica Fernández1, José M Cañas1, Vanessa Fernández1

  • 1Universidad Rey Juan Carlos, Móstoles, Spain.

Computational Intelligence and Neuroscience
|January 6, 2022
PubMed
Summary

This study introduces TrafficSensor, a deep learning system for real-time vehicle detection, classification, and tracking on highways. It utilizes YOLOv3/v4 networks and KLT tracking, performing effectively even in challenging conditions.

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

Field performance of MON 94637 and laboratory evidence supporting a Bt pyramiding strategy against key lepidopteran pests in Brazil.

Pest management science·2026
Same author

An Experiment in Personalized Shopping for Optimal Health, with Integration of Nutrigenetics and Gut Microbiome Information.

Nutrients·2026
Same author

Patient-targeted smartphone applications for pain management: A review of brazilian app markets.

Brazilian journal of physical therapy·2025
Same author

Efficacy of cognitive functional therapy for pain intensity and disability in patients with non-specific chronic low back pain: a randomised sham-controlled trial.

British journal of sports medicine·2025
Same author

'Despite the Pain, I Keep Moving Forward': A Qualitative Study on Brazilian Older Adults' Experiences With Chronic Low Back Pain.

Musculoskeletal care·2025
Same author

The Black and African American Connections to Parkinson's Disease (BLAAC PD) study protocol.

BMC neurology·2024

Area of Science:

  • Computer Vision
  • Deep Learning
  • Traffic Engineering

Background:

  • Real-time vehicle monitoring is crucial for traffic management and infrastructure planning.
  • Advances in neural networks have significantly improved object detection and classification capabilities.
  • Existing methods face challenges with poor lighting, weather, and low-resolution traffic images.

Purpose of the Study:

  • To develop and validate TrafficSensor, a deep learning system for automatic vehicle detection, classification, and tracking on highways.
  • To create and utilize a novel traffic image dataset encompassing diverse and challenging real-world conditions.
  • To evaluate the performance of different neural network models for vehicle detection and classification.

Main Methods:

  • Developed a two-module system: vehicle detection/classification and vehicle tracking.

More Related Videos

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K

Related Experiment Videos

Last Updated: Oct 7, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

674
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K
  • Employed YOLOv3 and YOLOv4 neural networks trained on a new, diverse traffic dataset.
  • Integrated a spatial association algorithm with the Kanade-Lucas-Tomasi (KLT) tracker for robust vehicle tracking.

    • Main Results:

    • The YOLOv3/v4-based network demonstrated high effectiveness in vehicle detection and classification.
    • The combined tracking module successfully followed vehicles in challenging traffic videos.
    • The TrafficSensor system achieved real-time performance in detecting, tracking, and classifying highway vehicles.

    Conclusions:

    • TrafficSensor provides an effective deep learning solution for real-time highway vehicle monitoring.
    • The system's ability to handle adverse conditions highlights the robustness of the developed models and dataset.
    • This research contributes to advancing intelligent transportation systems through accurate and efficient vehicle analysis.