MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641123501 A) filed by The. Principal, Sns College Of Technology on October 15, 2026, for Ai-Based Multimodel Lost-And-Found Item Matching System.
Inventors include Daze Thomas; Dr. M. Shobana; Dr. A. Sumithra; V. Vaishnavee; M. Nithish; B. Praneeth; T. Praveen; and N. Ragul.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention relates to an AI-based multimodal system and method for intelligent lost-and found item matching in smart campuses. The system centralizes the reporting, searching, and management of lost-item and found-item reports across educational institutions. Users submit item details, including the item name, category, description, image, location, date, and time. The system stores this information and identifies possible matches between lost-item and found-item reports. The system matches items using a multimodal approach. SBERT compares the semantic similarity Df textual descriptions, and CLIP compares the visual features of uploaded images. The system also evaluates location and time during matching. It combines these inputs into a single similarity score and ranks potential matches by that score. The system also explains matches and lets users verify them. It shows users which factors contributed to a suggested match, and users confirm or reject the result. When the system identifies a potential match, it sends a notification. Administrators manage users, reports, matches, and system records from the same platform. The invention combines multimodal AT matching, similarity scoring, match ranking, verification, notifications, and centralized data management. This reduces manual comparison and creates a more organized, efficient approach to lost-and-found management in smart campus environments.
Disclaimer: Curated by HT Syndication.
