MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202621029350 A) filed by Marathwada Mitra Mandal's College Of Engineering, Pune, Maharashtra, on March 12, for 'generation of solar potential map by adopting deep learning based footprint extraction.'
Inventor(s) include Darshan Joshi; Dr. Swati Shekapure; and Dr. Shaialja B. Jadhav.
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 and method for automated generation of solar potential maps by extracting suitable footprints (rooftops and open terrains) from high-resolution satellite imagery using deep learning models, followed by solar irradiance estimation and mapping. The system comprises of acquiring satellite datasets (e.g., Sentinel-2), pre-processing the imagery, applying convolutional neural networks such as U-Net and for precise footprint segmentation considering neighbouring terrain influences (buildings, vegetation, ridges), performing diurnal solar irradiance modeling with data sources like SOLCAST including azimuth, inclination, shading, tilt, and orientation corrections over temporal periods, aggregating radiation values to estimate potential per footprint, and generating interactive GIS-based visualizations categorizing zones by solar viability. This approach overcomes limitations of manual and traditional methods by providing scalable, high-accuracy identification of optimal solar installation sites. Principal use of the system includes aiding urban planners, solar providers, governments, and environmental agencies in sustainable energy planning and smart city development."
Disclaimer: Curated by HT Syndication.
