MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641008968 A) filed by Mohan Babu University; and Dr. K Reddy Madhavi, Tirupati, Andhra Pradesh, on Jan. 29, for 'ai-based automated medicinal plant recognition system using deep learning architectures.'
Inventor(s) include Dr. K Reddy Madhavi; Dr. B. Narendra Kumar Rao; Dr. V. Jalaja; and Dr. N Padmaja.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The present invention provides an automated, AI-based system for the correct identification of medicinal plant species using deep learning architectures. Multiple neural network models, including CNNs, VGG19, ResNet50, and EfficientNet, collectively form the system to extract visually complex features from plant leaf images captured in natural environments. A preprocessing module is provided to resize, normalize, and augment the data to improve quality and enhance model generalization. The preprocessed image goes through the deep learning inference engine for classification, generating probability scores for each plant species through a softmax-based classification module. Consequently, this would have very high accuracy even under varying lighting conditions, backgrounds, and orientations of leaves, thus enabling reliable real-time identifications on mobile devices, offline platforms, and embedded systems. Target agricultural, pharmaceutical, botanical, and conservation applications, the present invention can be used by users to identify the species, optionally along with some botanical information and its medicinal usage. This invention enhances plant authentication manifold and reduces the dependency on expert botanists while supporting scalability for deployment at research, industry, and field environments."
Disclaimer: Curated by HT Syndication.
