Traffic violations and related accidents, particularly among motorcyclists, are escalating in developing nations such as India. To address this issue, we developed a machine learning-based traffic surveillance system using the YOLO11 architecture. The system employs deep learning-based object detection to identify helmet usage and extract license plate information, enabling automated enforcement of helmet compliance regulations. YOLO11 was chosen for its better real-time detection capabilities, high precision, and efficiency in processing high-resolution imagery. The model was trained on a 260MB dataset containing annotated images, achieving an mAP50 of 91.4% in helmet detection and 85.2% in license plate recognition during validation. By automating traffic monitoring and minimizing human intervention, this approach enhances enforcement efficiency and contributes to improved road safety by ensuring accurate and reliable identification of non-compliant motorcyclists.