In recent years, Unmanned Aerial Vehicles (UAVs) and artificial intelligence (AI) have become game-changers in transportation engineering, providing new ways to monitor and control traffic. In this study, a quadcopter UAV- the ideaForge Q6 employed to capture high-resolution videos of traffic conditions at roundabouts in Prayagraj, India. The videos were then analyzed using the AI-based DataFromSky software, which automatically detects vehicles, classifies them, tracks their paths, and measures their speeds. Using georeferencing techniques, simple pixel data from the videos was converted into precise real-world measurements, facilitating detailed analysis of vehicle movements and identification of critical congestion points. The analysis revealed distinct patterns in traffic flow, deficiencies in lane discipline, and occurrences of speeding, all of which have significant implications for enhancing the safety and efficiency of roundabout operations. Heatmaps were generated to visualize speed variation and identify potential black spots with high conflict intensity. Additionally, trajectory data were used to examine traffic rule violations, including over-speeding and improper lane usage. The findings reveal critical flow patterns, behavioral deviations, and localized safety concerns. On the basis of these results, it is suggested that drone data be integrated into existing traffic management systems, install additional lanes in congested areas, and develop standard operating procedures for data retrieval. On the whole, this work shows that drones and AI can make a city's traffic network smart, safe, and effective.