MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123050 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 5, 2025, for 'deep learning system for heterogeneous traffic monitoring.'
Inventor(s) include Dr. S. Vasantha Kumar; Mr. Aditya Verma; Mr. Adrian L Synrem; and Ms. Kristina Shrestha.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure provides a deep learning system for heterogeneous traffic monitoring. The system includes a video input module receiving traffic video data from surveillance cameras, a YOLOv8 object detection module trained on twenty-one vehicle classes including auto-rickshaws and human haulers. The YOLOv8 module detects and classifies vehicles through a backbone network with convolutional layers and C2f blocks, a head network with concatenation blocks and upsampling layers, and output detection blocks generating bounding boxes and class predictions at multiple scales. A DeepSORT tracking module assigns unique tracking identifiers to detected vehicles across consecutive frames using Kalman filtering for motion prediction and appearance descriptors for vehicle matching, employing Mahalanobis distance calculations and Hungarian algorithm assignment. A counting module defines counting boundaries and increments vehicle counts when tracked vehicles cross boundaries. A flow rate calculation module computes traffic flow rates."
Disclaimer: Curated by HT Syndication.
