MUMBAI, India, Sept. 7 -- Intellectual Property India has published a patent application (202641104671 A) filed by Amrita Vishwa Vidyapeetham on August 31, 2026, for Biometric Identification System.
Inventors include Bhattacharya, Debanjali; Pati, Peeta Basa; Tripurari, Sameer Kumar Narasimha; Ranga Veera Manikanta, Deekshith Munuganti; and Poripireddi, Karthik.
The application for the patent was published on September 04, 2026, under issue no. 36/2026.
Abstract: Embodiments of the present disclosure relate to a system (100) and a method (500) of biometric identification. The system (100) receives multi-channel resting-state electroencephalography signals from a multi-channel EEG recording device (104), segments the received signals into fixed-duration EEG segments, and generates Continuous Wavelet Transform scalograms representing time-frequency characteristics of neural activity. The system (100) compresses the generated scalograms into latent feature patches using a convolutional autoencoder and spatially arranges the latent feature patches according to a predefined scalp electrode layout to construct two-dimensional cortical latent scalogram maps. The system (100) classifies the cortical latent scalogram maps using a residual neural network to generate segment- level biometric identities, evaluates prediction consistency, computes identification confidence, aggregates the predicted identities through segment-level majority voting, and determines a final authenticated subject identity for secure biometric identification.
Disclaimer: Curated by HT Syndication.
