MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641124097 A) filed by Amrita Vishwa Vidyapeetham on October 16, 2026, for System And Method For Authenticity Classification And Protection Of Digital Images.
Inventors include Muthurasu, Gayathri; Nagasubramanian, Aishwarya; Muralikrishnan, Mukesh; and Padmanapan, Deva Dharshini.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present disclosure relates to the field of image processing and machine-learning-based image analysis and provides a computing device (104) and a method (300) for enabling authenticity classification and protection of digital images. The computing device (104) stores in a memory (108) a trained image-authenticity classification model (110) that comprises model parameters generated using a training dataset, the training dataset including authentic images and artificial intelligence (AI)- generated images. The computing device (104) accesses an input image, determines, based on the trained image-authenticity classification model (110), a classification prediction indicative of whether the input image is an authentic image or an AI-generated image and selectively outputs at least one of an authentic image or a protected image based on the classification prediction indication. The protected image comprises a blurred image representation of the input image, the blurred image representation further comprising a visible watermark embedded in the blurred image representation.
Disclaimer: Curated by HT Syndication.
