MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621095655 A) filed by Dr. Saloni Niranjan Shah; Prof. Pooja Sambhaji Thengal; Prof. Dhanashri Rohidas Londhe; Prof. Seema Vilas Bhujade; Archana Yuvraj Jagtap; and Nikeeta Shamrao Shendkar on August 07, 2026, for Retina Authentication Security For Application..
Inventors include Dr. Saloni Niranjan Shah; Prof. Pooja Sambhaji Thengal; Prof. Dhanashri Rohidas Londhe; Prof. Seema Vilas Bhujade; Archana Yuvraj Jagtap; and Nikeeta Shamrao Shendkar.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: ABSTRACT [505] Reliable and tamper-resistant user authentication constitutes a foundational requirement for secure access to sensitive digital resources including banking systems, confidential data repositories, and defense-grade information networks; however, conventional authentication mechanisms such as passwords, personal identification numbers, and identity cards remain inherently vulnerable to theft, duplication, social engineering, and unauthorized replication, thereby exposing protected systems to unauthorized access. [510] Existing biometric authentication systems exhibit critical limitations including reliance on singular biometric modalities susceptible to spoofing through photographs, masks, or replay attacks, insufficient integration of internal ocular vasculature features that are substantially more resistant to forgery than external biometric traits, absence of real-time, camera-based retina isolation and matching pipelines suitable for deployment on standard consumer-grade computing hardware, and lack of combined multi-modal verification integrating both facial and retinal biometric evidence within a unified authentication decision. [515] The integration of live camera-based image acquisition, computer vision based retina region isolation, machine learning driven feature extraction, and combined facial-retinal pattern matching presents transformative opportunities for developing a web-deployable, real-time biometric authentication capability that is significantly more resistant to unauthorized access than traditional knowledge-based or single-modality authentication approaches. [520] The present invention describes a Retina and Facial Biometric Based Authentication Security System and Method comprising an image acquisition module that captures live facial and ocular imagery via webcam; a retina region isolation module that localizes and extracts the retinal vasculature region from the captured ocular image using image processing techniques; a feature extraction module that derives discriminative retinal vascular pattern descriptors using machine learning techniques; a facial recognition fusion module that independently verifies facial identity features; a pattern matching and authentication module that compares extracted features against enrolled reference templates stored in a secure database; and an anti-spoofing and liveness verification module that guards against fraudulent presentation attacks. [525] Comprehensive evaluation of the system across a benchmark dataset of enrolled user sessions demonstrated retina pattern matching accuracy of 96.2 ± 1.3 percent, combined facial-retinal fusion authentication accuracy of 98.1 ± 0.9 percent, false acceptance rate of under 0.4 percent, false rejection rate of under 1.8 percent, mean end-to-end authentication latency of under 900 milliseconds, and spoofing attempt detection accuracy of 94.6 ± 1.7 percent across simulated presentation-attack test scenarios. [530] These findings confirm that the Retina and Facial Biometric Based Authentication Security System and Method constitutes a significant advancement in biometric access control technology, with deployment applicability spanning secure banking systems, confidential data access platforms, defense and government security networks, and enterprise identity management systems, thereby addressing the substantial unmet need for a highly secure, difficult-to-forge, and real-time deployable authentication solution.
Disclaimer: Curated by HT Syndication.
