MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202511124203 A) filed by Manipal University, Jaipur, Rajasthan, on Dec. 9, 2025, for 'system for geographically contextual crater detection.'
Inventor(s) include Aditri Paul; Ritabrata Chakraborty; and Dr. Sandeep Chaurasia.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure relates to a system for geographically contextual crater detection designed to improve the accuracy and reliability of crater-based landmark identification for planetary navigation. The system receives raw descent imagery along with terrain context data obtained from onboard sensors and geological maps. A Context-Infused Super-Resolution (CISR) module enhances the imagery by incorporating terrain-class embeddings to recover fine crater morphology from low-resolution frames. The super-resolved image is processed by a deep-learning-based crater-detection network equipped with a Self-Calibrating Attention Mechanism (SCAM) and a multi-scale feature fusion neck. An uncertainty-aware detection head outputs crater locations, classifications, and associated confidence values. The detected craters are transformed into a crater-constellation graph, which is matched with a pre-computed onboard database using a Graph Attention Network-based relational matching module. A matched crater pattern is supplied to an adaptive extended kalman filter that integrates visual detections with inertial sensor data and adjusts measurement weights."
Disclaimer: Curated by HT Syndication.
