MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641099638 A) filed by Srm University-Ap on August 18, 2026, for System And Method For Land Use And Land Cover Classification Using Transition-Aware Contrastive Loss Framework.
Inventors include Dr. Achala Shakya; and Sri Harsha Vardhan Gogisetty.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: The present invention provides a system (100) and method (300) for multi-temporal land use and land cover (LULC) classification from satellite imagery. The system (100) comprises a computing unit (200) executing an image acquisition module, an image pre- processing module, a feature extraction module, a machine learning (ML) model configuration module, a spatiotemporal transition- aware contrastive loss (STAC-Loss) framework combining transition-weighted cross-entropy (TWCE), a spatial neighborhood consistency penalty (SNCP), and a class rarity contrastive term (CRCT) into a single optimization pass, a training module, a classification generation module and a performance evaluation module. The system (100) and method (300) together addresses dynamic temporal transitions, spatial neighborhood coherence, and class imbalance without modifying the internal architecture of the underlying deep neural network. Figure of Abstract –
Disclaimer: Curated by HT Syndication.
