MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641100696 A) filed by Jeppiaar Institute Of Technology on August 20, 2026, for Adaptive Biometric Privacy Preservation Framework For Real-Time Protection Against Unauthorized Faci.
Inventors include Ms Vishnu Prabha B; Ajeeth K; Dhanasekaran C; Kaviraj Jk; Nishant Nk; Gopika Arasi V; Daniell A; and Mugilan V.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: 6. ABSTRACT OF THE INVENTION (to be given along with complete specification on separate page) The present project proposes a browser-based privacy protection framework designed to mitigate the unauthorized collection and exploitation of users facial biometric information from images shared online. As facial recognition technologies increasingly rely on large-scale publicly available image datasets, users face growing risks of non-consensual biometric profiling, identity tracking, and unauthorized facial recognition. The proposed system integrates a browser extension with computer vision and adversarial perturbation techniques to protect facial images before they are uploaded to online platforms. The system automatically detects image upload activities and identifies facial regions within the selected image. Based on the detected facial features, imperceptible adversarial perturbations are generated and applied using techniques such as FGSM or PGD, while maintaining the visual quality of the image for human observers. The protected image is then substituted for the original image before the upload is completed. The framework also incorporates image-quality assessment using metrics such as PSNR and SSIM and evaluates the effectiveness of the protection against facial recognition models. By providing automated, real-time, and user-controlled protection at the browser level, the proposed system aims to reduce the risk of unethical biometric data scraping while preserving normal image-sharing functionality. The framework can be applied across various online environments, including social media platforms, professional networking websites, image-sharing services, and other web-based applications where users upload personal photographs.
Disclaimer: Curated by HT Syndication.
