Intellectual Property India Publishes Patent Application for '3D Human Biometric Multimodal Trait Identification System Using Deep Learning Model With Enhanced Anti-Spoofing And Illumination Invariance' Filed by Nitte Meenakshi Institute Of Technology, Nitte; Ms. Divya G; Dr. H C Nagaraj; and Dr. Mamatha A S

MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641051251 A) filed by Nitte Meenakshi Institute Of Technology, Nitte; Ms. Divya G; Dr. H C Nagaraj; and Dr. Mamatha A S, Bangalore, Karnataka, on April 22, for '3d human biometric multimodal trait identification system using deep learning model with enhanced anti-spoofing and illumination invariance.'

Inventor(s) include Nitte Meenakshi Institute Technology, Nitte; Ms. Divya G; Dr. H C Nagaraj; and Dr. Mamatha A S.

The application for the patent was published on May 1, under issue no. 18/2026.

According to the abstract released by the Intellectual Property India: "The present invention relates to a multimodal biometric authentication system that integrates both 2D and 3D biometric traits for enhanced security, robustness, and recognition performance. The system comprises a Data Acquisition Layer with 3D imaging sensors using structured light scanners and 2D imaging sensors for capturing hand surface geometry, finger knuckle patterns, and palm geometry; a Preprocessing 295 Layer for illumination correction, mesh smoothing, and depth normalization; a Deep Learning Feature Extraction Layer employing 3D CNN and 2D CNN for extracting geometric descriptors and texture patterns; a Multimodal Fusion Layer providing feature-level, score-level, and decision-level fusion; and a Matching and Classification Layer using SVM classifiers and NNDR matching algorithms. 300 The system utilizes 3D biometric data to overcome illumination, pose, aging, and abrasion-related variations typically faced by 2D biometric systems. By fusing texture-rich 2D information with depth-based 3D features using deep learning models, the system achieves high discrimination and improved reliability. The present invention provides anti-spoofing capability through 3D depth verification, reduces False Acceptance Rate 305 (FAR) and False Rejection Rate (FRR), achieves lower Equal Error Rate (EER), and ensures scalable, robust biometric authentication suitable for high-security applications including access control, banking, and identity management systems."

Disclaimer: Curated by HT Syndication.