MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641048242 A) filed by Gavini Venkata Leela Kumari; Bantupalii Kishor Babu; Rayapudi Hrsha Vardhani; Prajapathi Ravi; and Paravathaneni Sai Venkata Rajesh, Baptala, Andhra Pradesh, on April 15, for 'fake news and image detection using text analysis and ai generated image identification.'
Inventor(s) include Gavini Venkata Leela Kumari; Bantupalii Kishor Babu; Rayapudi Hrsha Vardhani; Prajapathi Ravi; and Paravathaneni Sai Venkata Rajesh.
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: "We claim, a Multimodal Fake News and Image Detection System, a deep learning-based invention configured to analyze news content using both text and image inputs to accurately detect and classify misinformation. 1. A multimodal framework comprising a text analysis module configured to process news content using Natural Language Processing techniques, an image analysis module configured to detect manipulated, morphed, or AI-generated images, and a classification module configured to determine whether the news is real or fake. 2. The system as claimed in claim 1, wherein the text and image modules operate independently to generate separate authenticity results, enabling identification of whether misinformation originates from text, image, or both. 3. The system as claimed in claim 1, wherein outputs from the text and image modules are combined using a fusion mechanism to improve accuracy and produce a final prediction along with a confidence score. 4. The system as claimed in claim 1, wherein the system supports text-only, image-only, and combined inputs and is applicable for social media monitoring, news verification, and digital forensic applications."
Disclaimer: Curated by HT Syndication.
