MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641050396 A) filed by Ambika Annavarapu; and Surekha Borra, Guntur, Andhra Pradesh, on April 20, for 'adaptive pareto-optimized multi-objective denoising system for remote sensing analytics.'
Inventor(s) include Ambika Annavarapu; and Surekha Borra.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present invention is a task-aware denoising optimization framework for remote sensing imagery that balances numerous downstream analytics objectives utilizing Pareto-optimal decision modelling. Sensor noise, atmospheric distortions, and mission-specific aberrations routinely degrade remote sensing data collected from satellites, airborne vehicles, and unmanned aerial platforms. Conventional denoising approaches improve a single picture quality metric, which frequently degrades characteristics required for automated analytics like classification, segmentation, and object detection. The proposed approach employs a multi-objective optimization framework that dynamically assesses trade-offs between conflicting analytics objectives before selecting Pareto-optimal denoising settings to preserve task-critical features while reducing noise. The system improves analytics performance by combining noise characterization, scene context analysis, task-aware deep learning denoising, and adaptive feedback optimization. The invention improves remote sensing intelligence extraction, decision-making dependability, and real-time deployment in ground-based processing systems, cloud environments, and edge-enabled onboard platforms by allowing for the dynamic balancing of different analytics goals."
Disclaimer: Curated by HT Syndication.
