MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641109515 A) filed by Mahendra College Of Engineering Autonomous on September 11, 2026, for An Ai-Based System And Method For Context-Aware Staff Presence And Classroom Behavior Monitoring.
Inventors include Mrs. S. Meera; Dr. V. Ponniyinselvan; Monika K; Dhanushya P B; Malini R; Anbukodi J; Girimurugan K; Harish M; Deepak R; and Gowthamraja S.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: [065] The present invention discloses an AI-based system and method for context-aware staff presence and classroom behavior monitoring in an educational space. One or more vision sensors provide observations to an edge perception node that generates pseudonymous person tracks and extracts zone, pose, motion, dwell, continuity and scene-quality features. A staff-presence and role inference engine and a classroom-behavior state engine independently generate hypotheses that are accumulated by a temporal evidence fusion engine over adaptive time windows. An adaptive policy and threshold engine selects a bounded room-specific or activity-specific policy profile, and an event qualification manager generates a qualified event only after persistence, contextual compatibility and uncertainty conditions are satisfied. Staff presence may be established without requiring a continuous facial-identity match. Event-relevant evidence may be privacy-transformed and recorded with model, policy, confidence and evidence-digest information in a tamper-evident ledger. Verified feedback can modify local thresholds within administrator-approved bounds to reduce recurring false alarms while preserving controlled operation. Accompanied Drawing [FIGS. 1-2].
Disclaimer: Curated by HT Syndication.
