MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202541113559 A) filed by A Janani Prasanna; R Anupama; S Dhana Varshini; M Dhanabalan; and P Kowsalya, Coimbatore, Tamil Nadu, on Nov. 19, 2025, for 'multi-script ocr and brahmi categorizer.'
Inventor(s) include A Janani Prasanna; R Anupama; S Dhana Varshini; M Dhanabalan; and P Kowsalya.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "Our novel system presents a deep learning-based method for precisely digitizing, identifying, and classifying multilingual scripts, including contemporary languages and ancient characters derived from Brahmi. The framework uses a dual-network design that consists of a recognition module that converts text regions into machine-readable digital text and a detection module that identifies and segments text regions. To improve clarity and feature extraction, raw Jocument or inscription images are enhanced, normalized, and converted into ideal input formats usmg sophisticated 1m age preparation techniques. In addition to classifying the identified scripts into families like Tamil, Devanagari, and Gujarati, the system can recognize printed, handwritten, and natural scene text. It achieves robust script categorization and good transcription accuracy through iterative refining. The system offers an effective, scalable, and automated technique for script recognition and digital archiving, and it is specifically designed for multilingual digitization, research, and heritage preservation."
Disclaimer: Curated by HT Syndication.
