MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641086325 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And Technology on July 14, 2026, for An Ai-Based Traffic Violation Detection And Automated Enforcement System Using Computer Vision, License Plate Recognition, And Real-Time Notification Mechanisms.

Inventors include Mr. Peddarapu Ramakrishna; Dr. P. Bharath Kumar Chowdary; Mrs. Ch. Sandhya Rani; and Mrs. K. Bhagya Rekha.

The application for the patent was published on July 17, 2026, under issue no. 29/2026.

Abstract: The increase in road accidents and violations such as triple riding, jumping the signal, and wrong lane riding shows that there is a requirement for a proper, mechanized system to ensure road safety. Manual enforcement procedures are typically time-consuming, random, and not suitable to be employed in most cities. To address this, we developed an online computer vision and machine learning-based traffic offense detection system to identify and report offenses from uploaded videos. The system is developed on Flask based on pre-trained Roboflow traffic sign and wrong-way detection, YOLOv7 car detection, and utilizes PaddleOCR integrated with Google Gemini for car number plates extraction from video frames. It then uses fuzzy string matching to match the extracted text with a list of known violators for absolute identification. In the event of a confirmed violation, a mail notification is sent to the user with all the details. SQLite and SMTP are used for user registration, authentication, as well as notification securely. Video uploaded is tagged, processed, and resized into mp4 so that it's web browser compatible. The model performed with 92.4% helmet detection accuracy, 88.7% accuracy in identification of triple riding, 90.3% accuracy in detecting signal violations, and 93.6% character-level accuracy for number plate recognition. The system offers an economical and scalable solution for traffic rule automation, which will assist in encouraging improved road discipline and public safety.

Disclaimer: Curated by HT Syndication.