MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202621101066 A) filed by Dr. D. Y. Patil Institute Of Technology, Pimpri, Pune on August 21, 2026, for Smart Helmet Detection And Automated Traffic Violation System Using Machine Learning.
Inventors include Ms. Yutika Patil; Mr. Sahil Selukar; Mr. Tejas Parmar; Dr. Shweta Deshmukh; and Dr. Vikash K. Agrawal.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: The invention discloses a machine learning-based system for real-time detection of helmetless two-wheeler riders and automated enforcement of helmet-wearing traffic regulations. The system employs a cascaded pair of YOLOv5-based deep learning models: a motorcycle detection model (YOLOv5-MD) that localises motorcycles within video frames captured by IP cameras installed at traffic junctions, and a helmet detection model (YOLOv5-HD) that evaluates the cropped rider region to classify helmet compliance. Upon detection of a violation, an Optical Character Recognition (OCR) module extracts the vehicle's registration number from the captured image, which is used to query a vehicle-owner database and retrieve the registered owner's contact details. The system automatically generates and dispatches a fine notification, including the applicable penalty amount as prescribed under the Motor Vehicles Act, 1988, to the violator via SMS and/or email, without requiring manual intervention by traffic police personnel. Each detected violation, together with its image evidence, timestamp, registration number, and owner details, is stored in a central database, enabling generation of violation reports and supporting long-term road-safety monitoring. By cascading motorcycle detection with a dedicated helmet- detection stage operating only on relevant cropped patches, rather than applying a single detector across an entire scene, the system improves detection accuracy, reduces false positives, and remains adaptable to varying camera placements and traffic conditions. The invention offers a significant improvement over existing manual enforcement methods and standalone detection or OCR systems by providing a complete, closed-loop, automated pipeline linking real-time detection, vehicle identification, and enforcement, thereby improving compliance with helmet-wearing laws, reducing traffic accidents and fatalities, and easing the enforcement burden on traffic police departments.
Disclaimer: Curated by HT Syndication.
