In today’s transportation landscape, two-wheelers have become the dominant choice despite their inherent risks due to minimal protection. India alone saw an average of 304 daily fatalities in 2020 from accidents involving unhelmeted riders, underscoring the need for proactive safety measures. Though governments have imposed penalties for non-compliance, traditional enforcement methods suffer from limitations like manual surveillance, limited coverage, and high costs. To address these challenges, we propose a novel aerial view solution for automating helmet detection to enforce regulations effectively. We introduce SkyGuard, a drone-based multi-head YOLOv5 with Convolutional Block Attention Module (CBAM) followed by a transformer block (C3TR) for improved detection. By employing a semi-supervised machine learning technique, we ensured a balanced representation. While shrinking human labor and improving accuracy, SkyGuard surpasses all the standard state-of-the-art detection models with a mAP:50, precision and recall rates of 98.5%, 96.6% and 94.7% on the OSF Helmet Dataset, respectively.