MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641048252 A) filed by National Institute Of Technology, Warangal, Telangana, on April 15, for 'a hardware-integrated intelligent system for real-time detection of grape leaf diseases using a convolutional capsule network.'
Inventor(s) include Mr. Banoth Prasad; and Prof. Raju Bhukya.
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: "The present invention discloses a hardware-integrated intelligent system for real-time detection of grape leaf diseases using a convolutional capsule network. The system comprises an image acquisition module, preprocessing module, convolutional neural network, capsule network, classification module, and output module. Images are resized to 128 128 and augmented using rotation, scaling, flipping, and gamma correction techniques. Additional convolutional layers are introduced before the primary capsule layer to reduce capsule complexity and accelerate routing. The system preserves spatial relationships and achieves classification accuracy of approximately 99.12%. The invention reduces computational overhead and improves detection efficiency. It is suitable for real-time agricultural applications and enables automated disease monitoring."
Disclaimer: Curated by HT Syndication.
