MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641108145 A) filed by Pragati Engineering College on September 09, 2026, for Intelligent Multi-Modal Edge Vision Framework For Railway Track Obstacle Detection With Explainable Decision.
Inventors include Dr. G. Naresh; Ms. Rubia Tasneem; Mrs. P. Sandhya; and Mr. Chembrolu Satish.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The present invention provides an Intelligent Multi-Modal Edge Vision Framework for Railway Track Obstacle Detection with Explainable Decision Support to boost railway operational safety in real time by detecting obstacles during the operation of the railway. It combines various sensors such as RGB cameras, thermal cameras, infrared cameras and environmental sensors and continuously monitors the railway tracks in multi-modalities under various weather and lighting conditions. The acquired sensor information is synchronized and undergoes adaptive pre-processing of the images, noise reduction, compensation of the environmental conditions, and normalization of the data, and finally processes them by adaptive multi-modal sensor fusion to obtain a reliable image of the environment monitored by the railway.The fused data are fed into an edge-based AI system capable of real-time obstacle detection and identification without relying on cloud systems, reducing communication latency and allowing for timely safety interventions. The framework can detect various track obstacles, such as humans, animals, vehicles, fallen trees, rocks, debris, damaged rails, and other dangerous objects. A context-aware risk assessment module analyzes parameters of obstacles and operational and environmental parameters to assess the severity of risk. Explainable decision support mechanism provides transparent and interpretable decisions for each detection, classification, and safety recommendation, which boosts the operator's confidence and supports informed decision making in safety-critical situations.Depending on the risk assessment, an intelligent safety response and communication module automatically triggers warning messages, speed reduction suggestions, emergency brake suggestions and maintenance alerts to railway operators and control centers. Another aspect of the framework is operational event and decision logs which are stored for future analysis and to improve the system. The invention provides a holistic solution for railway track monitoring based on a single solution architecture, adaptive multi-modal sensing, edge artificial intelligence, explainable decision support and intelligent safety response, which is accurate, reliable, transparent and efficient, and can significantly mitigate the risk of accidents, enhance the safety, reliability and efficiency of modern railway transportation systems.
Disclaimer: Curated by HT Syndication.
