Intellectual Property India Publishes Patent Application for 'Tamil Handwritten Text Recognition And Translation System Using Convolutional Recurrent Neural Networks' Filed by Vellore Institute Of Technology

MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122206 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 4, 2025, for 'tamil handwritten text recognition and translation system using convolutional recurrent neural networks.'

Inventor(s) include Dr. R. Manjula; Tarun Sri Darshan Balachandran; Pranav Veerendra Kumar; and Aditya Nair.

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 present disclosure provides a system for Tamil handwritten text recognition and English translation. The system includes a preprocessing module for input image processing, a spatial transformer network (STN) for geometric normalization to correct distortions, a convolutional neural network for spatial feature extraction, a recurrent neural network for sequential processing to model text dependencies, a connectionist temporal classification decoder for generating recognized Tamil text, and a neural machine translation module for English translation. The STN applies thin plate spline transformation through TPS-Spatial Transformer (336 256) for geometric alignment. The CNN employs a residual network feature extractor with multiple residual blocks. The RNN utilizes bidirectional LSTM layers (BiLSTM 1, BiLSTM 2) to capture contextual dependencies in both directions for accurate Tamil script recognition."

Disclaimer: Curated by HT Syndication.