MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641097938 A) filed by Sri Eshwar College Of Engineering on August 12, 2026, for Agri Nova - Al Powered Smart Agriculture Ecosystem.
Inventors include Mr. Gowrisanker M; S G Lalith Prabu; and V Hari Balaji.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: AGRI NOVA is a digital system that aims to help Indian farmers find plant diseases in their crops using artificial intelligence. The system lets users find plant diseases by taking photos of crops or leaves with their phones. The app uses a Next.js and TypeScript end. The back end is built with Python and FastAPI. It also includes machine learning services. Uses Supabase for the database. The main feature of the app is a machine learning process that finds diseases by looking at pictures of crops and leaves. The current back end works with TensorFlow and Keras models and PyTorch models. It uses models like MobileNetV2 and MobileNetV3 so the system can be used in real farming situations. In addition the app has an AI assistant that helps farmers with questions about crop diseases and how to find them. The future of the project has plans for improvements. Some of these improvements include adding real-world data sets like PlantVillage and PlantDoc to make disease detection better in life. The plan also includes using advanced data augmentation methods with Albumentations. The model architecture will be changed to EfficientNetB3 or ConvNeXt. Also the future plan includes combining the training and inferencing parts into one system using TensorFlow. On the app can be made better by using YOLOv8 or YOLOv11 for object detection. This would allow the app to highlight the parts of the leaf that are infected with a box. The idea behind AGRI NOVA is to bring artificial intelligence and computer vision with a chat assistant to give farmers a useful tool, for finding crop diseases early.
Disclaimer: Curated by HT Syndication.
