Discriminative Image Feature Extraction for Traffic Sign Detection in Road Inspection

Project Code :TCMAPY2104

Objective

This project uses YOLOv11 and YOLOv12 deep learning models for efficient traffic sign detection in road inspections. Trained on a Roboflow dataset, the system accurately identifies signs like Hazard, No Entry, Speed Limits, Stop, Traffic Lights, and Yield. A user-friendly Streamlit interface allows users to register, upload images, and receive predictions. It also includes live streaming for real-time detection. This system enhances traffic safety and operational efficiency by automating road monitoring.

Abstract

This project aims to enhance road inspection efficiency by utilizing advanced deep learning models, YOLOv11 and YOLOv12, for traffic sign detection. Trained on a robust dataset from Roboflow, the system can accurately identify various traffic sign classes such as Hazard, No Entry, Speed Limits, Stop, Traffic Lights, and Yield signs. YOLOv11 and YOLOv12 algorithms are employed for feature extraction and classification, ensuring high accuracy in detection. The project integrates a user-friendly interface built with Streamlit, allowing users to register, upload images, and receive predictions for traffic sign detection. Additionally, the system includes live streaming functionality, providing real-time traffic sign detection. By automating road inspections, this system aims to improve traffic safety and operational efficiency, making road monitoring faster and more reliable.

Keywords:
Traffic Sign Detection, YOLOv11, YOLOv12, Deep Learning, Road Inspection, Streamlit, Image Processing, Prediction, Traffic Safety, Automated Inspection.

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.