Next-Generation Plant Management System and Disease Detection Using Advanced Deep Learning

Plant diseases significantly impact agriculture by reducing crop yield and overall harvest efficiency. This research presents a disease detection model integrated with a self-calibrated illumination enhancement model for improving the quality of the image under the low light conditions. For detection, MobileViT model is implemented with a Flutter-based mobile application, using TensorFlow Lite for real-time inference. For treating the plants, a remedy generation system is implemented using a Sequence-to-Sequence LSTM based Text Generation model. The application includes exclusive features which includes an AI chatbot powered by Dialogflow, plant identification through Plant.id API, community support for users interaction, an E-commerce store using Firebase and Smart Plant Manager is included to monitor plant health by combining GPS and weather data for an enhanced user experience. Finally, multilingual support is ensured for user through the MobileNMT model, ensuring accessibility for diverse users.