MUMBAI, India, Nov. 21 -- Intellectual Property India has published a patent application (202441038881 A) filed by Ankita Rishabh; and Timo Traurig, Bengaluru, Karnataka, on May 17, 2024, for 'biometric authentication methods based on multimodal fusion.'
Inventor(s) include Ankita Rishabh; and Timo Traurig.
The application for the patent was published on Nov. 21, under issue no. 47/2025.
According to the abstract released by the Intellectual Property India: "The present invention pertains to a novel method of biometric authentication based on multimodal fusion that can improve the accuracy, security, and convenience of biometric authentication. The said methods comprise the steps of: Acquiring biometric data from two or more biometric modalities, such as face, iris, fingerprint, voice, or gait, using suitable sensors or devices. Extracting biometric features from the acquired biometric data using various feature extraction techniques, such as local binary patterns, Gabor filters, deep neural networks, or wavelet transforms. Fusing the biometric features from different modalities using various fusion techniques, such as score-level fusion, decision-level fusion, feature-level fusion, or hybrid fusion. Matching the fused biometric features with the stored biometric templates using various matching algorithms, such as cosine similarity, Euclidean distance, Hamming distance, or support vector machines. Generating a matching score and comparing it with a predefined threshold to determine whether the person is authenticated or not. The methods of the present invention can be applied to various scenarios, such as online or offline authentication, verification or identification, single or multiple users, and static or dynamic biometrics. The said methods can also be adapted to different constraints, such as computational cost, communication bandwidth, storage capacity, or user preference."
Disclaimer: Curated by HT Syndication.
