MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202611076915 A) filed by Dit University on June 22, 2026, for “a Multi-Modal Anti-Spoof Face Recognition Attendance System”.
Inventor includes Dr. Garima Verma.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention relates to a multi-modal anti-spoof face recognition attendance system and method are disclosed. A biometric acquisition layer captures facial image frames, three-dimensional facial contour data, and facial temperature distribution data using an RGB camera, a depth sensor, and a thermal sensor. A processing and authentication layer performs face detection, feature extraction, template matching, depth contour validation, thermal consistency analysis, and eye-blink liveness verification including eye-aspect-ratio computation and a randomized blink challenge. A trust scoring and decision layer computes a composite trust score by weighted fusion of normalised facial recognition, depth consistency, thermal consistency, and liveness confidence values, and compares the composite trust score with a threshold dynamically adjusted based on ambient illumination, subject-to-sensor distance, and spoof attempt history. An attendance logging layer records attendance only when the composite trust score satisfies the threshold, logs failed attempts with a rejection reason, snapshot, and alert, and generates periodic attendance summaries. Fig 1 to 5
Disclaimer: Curated by HT Syndication.
