MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641109711 A) filed by Sathyabama Institute Of Science And Technology; Dr. G. Sukanya; Dr. P. Karthika; G Karthika Priya Dharshini; Ms. Leena Sivaguru; Dr A P Narmadha; Dr. V. R. Niveditha; and Ms. Devi. D on September 12, 2026, for Emergency Vehicle Recognition: A Cnn Approach For Efficient Classification.
Inventors include Dr. G. Sukanya; Dr. P. Karthika; G Karthika Priya Dharshini; Ms. Leena Sivaguru; Dr A P Narmadha; Dr. V. R. Niveditha; and Ms. Devi. D.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: Abstract: TITLE: EMERGENCY VEHICLE RECOGNITION: A CNN APPROACH FOR EFFICIENT CLASSIFICATION The rapid growth of urban populations and increasing traffic congestion have critically affected the response times of emergency vehicles ambulances, fire trucks, and police cars which are essential for life- saving interventions and public safety. Conventional traffic management systems are often incapable of detecting and prioritizing emergency vehicles in real time, leading to critical delays with potentially severe consequences. This project presents the Phase II implementation of Emergency Vehicle Recognition: A CNN Approach for Efficient Classification. Building upon the Phase I system design and literature review, Phase II encompasses the complete implementation, training, deployment, and testing of the Xception-based Convolutional Neural Network (CNN) model integrated with a Flask web application. The model employs transfer learning from ImageNet pre-trained weights, achieving an overall classification accuracy of 95.6%, precision of 94.8%, recall of 96.2%, and F1-score of 95.5% on the held-out test dataset. The deployed system processes video frames in approximately 50 milliseconds per frame, enabling real-time emergency vehicle detection through a web-based interface. Upon detection, the system automatically triggers a green corridor mechanism that interfaces with traffic control infrastructure to prioritize emergency routes. Evaluation against benchmark models VGG16 and ResNet50 demonstrates superior accuracy and inference speed, contributing a lightweight, scalable, and deployable solution for smart city traffic ecosystems.
Disclaimer: Curated by HT Syndication.
