Related Experiment Video
Updated: Feb 18, 2026
14:55
Evaluating the Effect of Roadside Parking on a Dual-Direction Urban Street
Published on: January 20, 2023
4.4K
A video-based real-time adaptive vehicle-counting system for urban roads
Fei Liu1, Zhiyuan Zeng1, Rong Jiang2
1School of Hydropower and Information Engineering, Huazhong University of Science and Technology, Wuhan, China.
Plos One
|November 15, 2017
Summary
This study introduces an adaptive computer vision model for real-time vehicle counting in urban areas. The system achieves over 99% accuracy, aiding traffic management in developing nations.
Area of Science:
- Computer Vision
- Urban Planning
- Traffic Engineering
Background:
- Rapid urbanization in developing nations leads to significant traffic management challenges.
- Accurate, real-time traffic flow data is essential for effective urban mobility solutions.
- Existing traffic monitoring systems often lack precision and real-time capabilities.
Purpose of the Study:
- To develop an adaptive computer vision model for real-time vehicle counting on urban roads.
- To enhance urban traffic management systems with precise and reliable data.
- To introduce a robust method for monitoring real-time traffic congestion.
Main Methods:
- An automatic real-time background update algorithm for vehicle detection.
- An adaptive vehicle counting pattern utilizing virtual loop and detection line methods.
- A novel robust detection technique for assessing road section congestion.
Main Results:
- A functional prototype system was developed and deployed on an urban road.
- The system demonstrated robustness in real-world field scenarios.
- Real-time vehicle counting accuracy consistently exceeded 99%.
Conclusions:
- The proposed computer vision model provides a highly accurate and robust solution for real-time traffic monitoring.
- This technology can significantly improve urban traffic management, especially in developing countries.
- The system offers a reliable method for assessing traffic congestion states.
