کنفرانس مهندسی برق ایران

صفحه اصلی / سی و سومین کنفرانس بین المللی مهندسی برق

A Cost-Effective Solution for Traffic Sign Recognition and Geographic Localization Using a Monocular Camera

نویسندگان :

Mohadeseh Atyabi¹ Fardin Ayar² Mahdi Javanmardi³

1- Department of Computer Engineering Amirkabir University of Technology 2- Department of Computer Engineering Amirkabir University of Technology 3- Department of Computer Engineering Amirkabir University of Technology

کلمات کلیدی :

Traffic sign recognition،Object Detection،Machine Learning،Geographic Information System

چکیده :

Efficient management of traffic sign locations is critical for road safety and smooth traffic flow. This study presents an automated system that detects and localizes traffic signs using images captured by a mobile phone. The system fine-tunes a pre-trained Faster R-CNN object detection model on a dataset of Tehran's street view images to identify traffic signs. Assuming predefined traffic signs' real-world sizes, It then estimates the position of each sign relative to the camera using the Perspective-n-Point (PnP) algorithm. By combining the camera’s GPS coordinates and the estimated positions, the system calculates the traffic sign's precise geographical location and visualizes it on a map. The accuracy of the system is validated by comparing the generated map with real-world images, demonstrating a high level of alignment.