MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641099799 A) filed by Rajalakshmi Engineering College on August 18, 2026, for An Intelligent Deep Learning Framework For Real-Time Marine Oil Spill Identification And Classification.
Inventor includes Dr. J. Senthil Murugan.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: An intelligent deep learning framework is presented for continuous oceanic surveillance and real-time identification and classification of marine oil spills. The system operates through a multi-platform sensor networkincluding satellites, unmanned aerial vehicles, aircraft, and surface buoysto harvest multimodal spectral, optical, and radar imagery across marine environments. Incoming data streams undergo automated radiometric calibration, geometric alignment, and clutter suppression to eliminate oceanic wave noise, solar glint, and organic sheens while extracting key spatial and spectral features. These refined inputs are processed by a deep neural network engine combining convolutional models and vision transformers to differentiate authentic petroleum spills from false-positive look-alikes, such as algae blooms or low-wind areas. Concurrently, the neural engine classifies the specific oil typesuch as crude oil, diesel, or heavy fueland estimates physical attributes including slick thickness, total surface area, and volume. Predictions are rendered onto an interactive analytics dashboard displaying live geo-spatial map overlays, dynamic drift projections, and severity indicators. Upon event verification, an automated early warning system broadcasts encrypted alerts, geographic target coordinates, and tactical containment protocols directly to coastal authorities and maritime response fleets to minimize ecological impact.
Disclaimer: Curated by HT Syndication.
