International Journal of Scientific Research and Engineering Development

Road safety violations such as riding without helmets and triple riding are prevalent concerns in college campuses, posing significant risks to student safety. Traditional monitoring methods rely heavily on manual supervision by security personnel, which is inefficient, inconsistent, and unable to provide realtime detection. This paper presents an AI-Based Smart Motorcycle Safety Monitoring System designed specifically for campus environments. The system utilizes CCTV cameras to continuously capture realtime video of moving vehicles. The captured video is processed using OpenCV for frame extraction and image preprocessing. A deep learning-based object detection model, YOLO (You Only Look Once), is applied to detect motorcycles, identify helmet usage, and count the number of riders on each motorcycle. Upon detecting a violation such as no helmet or triple riding, the system automatically captures the violation frame and generates an alert with visual evidence. The system is implemented using Python with OpenCV, Ultralytics YOLO, and NumPy libraries. The proposed system eliminates continuous human monitoring, improves detection accuracy, and ensures efficient enforcement of campus traffic safety regulations.

Shifa Bilal Tamboli, Simeen Phiroj Mulani, Arman Tajuddin Shiakh,"AI Based Smart Motorcycle Safety Monitoring System for Campus Environments" International Journal of Scientific Research and Engineering Development, V9(2): Page(2571-2575) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.