MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641051653 A) filed by Neerupudi Mangala Deepika; and Dr. K. Sreerama Murthy, Hyderabad, Telangana, on April 23, for 'a scalable edge-deployable framework for intelligent plant leaf condition analysis using compressed feature encoding and multi-scale attention-based classification.'
Inventor(s) include Neerupudi Mangala Deepika; and Dr. K. Sreerama Murthy.
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: "Feature Encoding and Multi-Scale Attention-Based Classification The present disclosure relates to an intelligent plant health monitoring framework configured for automated analysis of plant leaf conditions using deep learning techniques. The framework comprises an image capture interface, a preprocessing engine adapted to standardize acquired images, a feature encoding module configured to generate compact latent representations through a convolutional encoding mechanism, and a classification engine employing a lightweight neural architecture integrated with a multi-scale attention mechanism. The system is configured to dynamically emphasize diagnostically relevant regions within plant leaf images while minimizing computational overhead through compressed-domain processing. A decision module categorizes plant conditions into multiple health states and generates output data including classification labels, confidence metrics, and crop condition indicators. The framework is further adapted for execution across distributed environments including edge devices, mobile platforms, and cloud infrastructures."
Disclaimer: Curated by HT Syndication.
