Integration of IoT and Machine Learning for Real-Time Plant Health Monitoring and Disease Detection System

Abstract

Abstract – Agricultural yield is highly dependent on timely disease management and optimal growing conditions. In mango cultivation, especially for the Alphonso variety, diseases such as anthracnose and rust cause significant damage. This paper introduces an IoT-based system that combines environmental monitoring with machine learning-driven leaf disease detection. Temperature, humidity, and soil moisture are tracked using DHT11 and soil moisture sensors interfaced with a NodeMCU ESP8266. This data is visualized on ThingSpeak. For disease diagnosis, a trained Convolutional Neural Network (CNN) model classifies mango leaves into healthy, rust-infected, or fungal-infected categories. The model is deployed via a Streamlit web application, offering users an intuitive interface for image upload and result display. The integrated system supports precision agriculture through timely alerts and remedies, reducing manual inspection and promoting sustainable farming. Keywords: Alphonso mango, Convolutional Neural Network (CNN), IoT-based monitoring, Leaf disease detection, Smart agriculture, Internet of Things (IoT).

Plant phenotyping relevance

マンゴー葉の画像から病害状態をCNNで推定する手法と、IoTセンサーを統合したモニタリングシステムが中心であり、植物の病害表現型を直接評価している。

Code and data availability

The paper describes a custom-curated mango leaf image dataset, a trained CNN (.h5), and Streamlit/NodeMCU code, but provides no public deposit, repository URL, or availability statement for any of them. The only URL given is the generic ThingSpeak documentation, which is a third-party platform reference, not a paper-­­

No evidence-backed public reproduction asset is currently recorded.