MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202621075904 A) filed by Symbiosis International Deemed University on June 18, 2026, for Deep Learning-Based Road Quality Detection And Monitoring System Using Real-Time Tracking And Gps Visualization.
Inventors include Ketan Agarwal; Saksham Chimdyalwar; Lokesh Polkam; and Dr. Bhupesh Kumar Dewangan.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: ABSTRACT DEEP LEARNING-BASED ROAD QUALITY DETECTION AND MONITORING SYSTEM USING REAL-TIME TRACKING AND GPS VISUALIZATION The present invention discloses an integrated deep learning-based road quality detection and monitoring system (100) for automated identification, classification, segmentation, and tracking of road surface defects. The system (100) comprises an input module (110) for receiving image or video data, a preprocessing module (120) for frame extraction and normalization, a detection module (130) employing a YOLOv8 deep learning architecture for real-time defect detection and classification, a segmentation module (140) for pixel-level defect boundary delineation through instance segmentation, a tracking module (150) for assigning unique identifiers to defects across video frames to eliminate duplicate counting, a scoring module (160) for computing composite road quality scores based on defect type, area, and severity, and a visualization module (170) providing a Streamlit-based web interface with GPS mapping, annotated output display, and QR code enabled mobile access. [
Disclaimer: Curated by HT Syndication.
