MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202611086397 A) filed by Chandigarh University on July 15, 2026, for Privacy-Preserving Edge Ai System For Real-Time Vandalism Detection And Evidence Authentication.
Inventors include Ayush Mishra; Inakshi Garg; Parveen Bajaj; Anshul Saxena; Sanjeet Kumar Chaudhari; and Vikas Thakur.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The present invention relates to a privacy-preserving edge AI system for real-time vandalism detection and secure evidence generation. The system comprises one or more cameras integrated with an edge computing device configured to locally process video streams without continuous cloud dependence. Motion detection and image pre-processing activate intelligent analysis only when required to optimize computational resources. The system employs object detection, semantic segmentation, action recognition, pose estimation, and temporal event fusion to accurately identify vandalism activities including graffiti spraying, scratching, sticker pasting, and deliberate property damage while minimizing false alarms. A damage assessment module quantifies the affected area and assigns severity levels. A privacy module anonymizes faces and vehicle number plates before storage or transmission. Digitally signed evidence packages containing anonymized video, metadata, timestamps, and severity scores are securely transmitted to authorized users. Human-in-the-loop verification and federated learning continuously enhance detection accuracy while preserving data privacy and evidence integrity.
Disclaimer: Curated by HT Syndication.
