EnviScan+: AI-driven plant identification and eco-system management for sustainable agriculture

Abstract

EnviScan+ is an interactive mobile and web application that facilitates plant recognition and preservation through AI-based image identification and a chatbot-oriented recommendation system. By pairing deep learning with real-time assistance, it enables its users to accurately categorize plants and benefit from personalized care recommendations. Built on a modern distributed architecture approach with a Spring Boot backend, a native Android version and a React-based web platform, EnviScan+ ensures user-friendly operation tailored to users’ needs. Its plant species recognition mechanism exploits convolutional neural networks (CNNs) trained on an extensive dataset for high accuracy, while the chatbot provides immediate explanations to plant species questions. Additional features such as automated watering reminders, multilingual support and a constantly updated plant database make plant care simpler and more acceptable. To ensure the safety and quality of the solution, the platform is rigorously evaluated using SonarQube analysis. By combining high technology with practical plant care, EnviScan+ crosses the threshold between botanical expertise and everyday users, bringing artificial intelligence-based plant maintenance within the reach of researchers, enthusiasts and agricultural specialists.