Artificial Intelligence - Computer Vision
AI Traffic Monitoring Application with Object Detection and Counting
CNN-based object detection that counts vehicles crossing a traffic signal and shows the number detected, by type, in a table. It uses a pretrained YOLO (You Only Look Once) model.
What it does
This AI application detects traffic objects such as cars, trucks, bicycles, motorcycles and pedestrians using computer vision, and counts them as they cross a traffic signal. It uses a pretrained YOLO (You Only Look Once) model for object detection in images, video frames or live camera feeds.
Testing
The application was tested on an NVIDIA Jetson Nano for live stream processing and on Ubuntu 24.04 for prerecorded videos.
Extensions
The project can be extended to other use cases, including:
- Parking lot occupancy detection
- Pedestrian flow monitoring
- Traffic anomaly detection (for example, driving against traffic)
- Real-time lane usage and congestion tracking
This version is implemented in Python and does not run in real time. I have also implemented a C++ version that does. It will be published once the installer is ready, as its dependencies currently cause installation issues across different Ubuntu versions.
For more details visit https://github.com/neoviki/vehiclecounter
References
- Ultralytics YOLOv8 documentation: https://docs.ultralytics.com/models/yolov8/
- Ultralytics GitHub repository: https://github.com/ultralytics/ultralytics
- Video by German Korb on Pexels: https://www.pexels.com/video/road-systems-in-montreal-canada-for-traffic-management-of-motor-vehicles-3727445/