MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202631017942 A) filed by Guru Nanak Institute Of Technology, Kolkata, West Bengal, on Feb. 18, for 'emotion recognition from handwriting using multimodal data fusion.'
Inventor(s) include Dipankar Basu; Nilanjana Adhikari; Ananjan Maiti; Prasansha Bairagi; and Barsha Pan.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "A new emotion recognition framework using multimodal handwriting analysis in analysing dynamic writing features (speed, pressure, and trajectory), spatial attributes (spacing, slant, and formation), and temporal signatures (pausing, rhythm). Advanced data fusion methods combine these modalities to form a new emotion recognition solution. The findings demonstrate the novel system achieving 92% emotion recognition accuracy using an ensemble deep neural network across 7 different emotion categories (joy, sadness, anger, fear, surprise, disgust, neutral) and a real-time response time 350 ms. The important contributions of the framework include: privacy-preserving processing on a device; adaptive personalization; and universal compatibility on generic digital writing devices and paper-based writing systems. The multimodal system significantly outperforms traditional single-modality methods by demonstrating accurate and reliable emotion inference for a cohort of diverse users under varying writing conditions, time-scales and environmental contexts. Future applications may exist for healthcare and healthcare monitoring, educational performance evaluation, human-computer interaction, and therapeutic intervention assistance."
Disclaimer: Curated by HT Syndication.
