MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122330 A) filed by Sai Vidya Institute Of Technology, Bengaluru, Karnataka, on Dec. 5, 2025, for 'geo vision plan-net: deep multi-class segmentation for smarter urban planning.'
Inventor(s) include Dr. T. G. Manjunath; Dr. Manjunath T N; Dr. Vikramathithan A C; Rekha Murthy; Jayshri; Deepika G S; R V Anusha; Harichandan G; Poornima P; and Sohan Cholaraju.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "This research proposes a deep learning-based multi-class semantic segmentation framework for high-resolution satellite imagery to support large-scale urban planning. The system leverages DeepLabV3++ with Atrous Spatial Pyramid Pooling (ASPP) and a robust preprocessing pipeline including augmentation, normalization, and tiling to extract fine-grained spatial features. The model classifies key urban land cover categories such as buildings, roads, vegetation, and water bodies, producing high precision segmentation maps suitable for GIS integration. Comprehensive evaluation using IoU, Precision, Recall, and F1-score demonstrates strong performance and cross regional generalization across varying resolutions and imaging conditions. The technical novelty of the framework lies in its unified pipeline combining multi-scale feature extraction, multi-class prediction, and post-processing refinement, enabling an accurate and scalable solution for automated urban analysis and sustainable city development."
Disclaimer: Curated by HT Syndication.
