Portfolio

Overview:

This project details the development of an AI-powered traffic surveillance system aimed at reducing speeding-related fatalities. The system is built using Edge AI for vehicle detection and speed analysis, a main platform for data aggregation, and a user form for public reporting. The product focuses on real-time vehicle detection and speed reporting, with automated notifications sent to relevant authorities when violations are detected.

Screenshots:

Key Features:

1. Edge Device:

• Vehicle Detection: Uses YOLO for fast real-time vehicle detection. A tracking ID is assigned to each vehicle, allowing continuous tracking through video frames.
• Speed Estimation: Tracks a vehicle’s movement across a pre-defined line and calculates its speed based on the time taken to cross the line.
• Car Counting: The system counts cars by tracking their movement as they cross a virtual line within the frame.
• Speeding Report: If a vehicle exceeds the speed limit by 20%, an automated email is sent. The email contains key information, including vehicle speed, time of detection, and the vehicle’s details.

2. Main Platform:

• Main Dashboard: The dashboard provides a user-friendly interface for monitoring the data collected by the edge devices and user forms. It features simple graphs, maps, and sortable data displays, helping users analyze trends such as the number of speeding incidents over time.
• Data Analysis: Operators can upload and analyze video data, extracting vehicle speed, count, and direction. Though real-time testing isn’t always feasible, this function simulates real-time analysis by processing uploaded video data.
• Reports and Tables: The platform provides detailed tables that display information on all detected vehicles, including speeding incidents and user-reported data. The system highlights vehicles with multiple speeding violations for further review.

3. User Form:

• Incident Reporting: End users can report traffic issues by submitting photos and videos via the user form. The system then analyzes these submissions using the algorithm to determine if any reported vehicles are flagged for speeding. If the vehicle is already registered as high-risk, an automatic report is generated for authorities.

Category:

Tags:

Links:

• GitHub
• YouTube