MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122164 A) filed by Mr. T. Prabu; Mr. P. Balasubramani; Ms. Sowmiya K; Mr. Jeniton S; Mr. K. Arun Patrick; Ms. Cowsalya T; Sanjay Kumar. S; Nisanth. U; Abisheak. R; and Sudhakar. V, Coimbatore, Tamil Nadu, on Dec. 4, 2025, for 'ai-powered ocr for translating handwritten documents into regional languages.'
Inventor(s) include Mr. T. Prabu; Mr. P. Balasubramani; Ms. Sowmiya K; Mr. Jeniton S; Mr. K. Arun Patrick; Ms. Cowsalya T; Sanjay Kumar. S; Nisanth. U; Abisheak. R; and Sudhakar. V.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The invention provides an AI-powered OCR system designed to digitize handwritten documents and translate the recognized text into regional languages. It integrates Convolutional Neural Networks (CNN) for feature extraction, Recurrent Neural Networks (RNN/LSTM) for sequence prediction, and Transformer-based models (such as TrOCR) for enhanced recognition and contextual translation.The system performs image preprocessing, including noise reduction, binarization, skew correction, and contrast enhancement, to improve OCR accuracy. Once text is extracted, it is processed by NLP and multilingual translation models to generate contextually accurate outputs in languages such as Tamil, Hindi, Telugu, Kannada, and English. The invention further includes a web-based interface that allows users to upload images or capture live photos using a camera, enabling real-time OCR and translation. The results can be saved, archived, or integrated into digital libraries. This unified pipeline significantly improves accuracy, multilingual accessibility, and preservation of handwritten cultural documents."
Disclaimer: Curated by HT Syndication.
