MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641051553 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on April 22, for 'a system for dual-encoder spatio-structural road degradation detection from satellite imagery.'
Inventor(s) include Ragavan K; Krithika Vivekanandan; and Kishan S.
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 relates to a system (100) for detecting infrastructure degradation from satellite imagery. The system (100) comprises an image acquisition module configured to retrieve satellite images based on geographic coordinates and a preprocessing module configured to generate a road region mask using colour-space transformation and morphological operations. A dual-branch neural network module processes RGB colour-texture features and edge-gradient features in parallel encoder branches, wherein edge features are generated using a Sobel operator. A feature fusion module employing channel attention mechanisms combines outputs of the encoder branches, and a decoder module generates a crack probability map and corresponding binary mask. An analysis module computes a degradation index based on crack width, density, connectivity, and temporal parameters. The system (100) operates as an integrated processing pipeline for infrastructure analysis."
Disclaimer: Curated by HT Syndication.
