AgroPulse: A Real-Time Field Intelligence System for Crop Disease Tracking and Notification System

Abstract

Agriculture faces significant challenges due to plant diseases, which directly affect crop yield, quality, and farmer income. Early detection of agricultural diseases is critical to prevent large-scale crop losses and reduce excessive use of pesticides. Traditional disease detection methods rely on manual inspection by farmers or agricultural experts, which is time-consuming, subjective, and often inaccurate, especially during early stages of infection. With the advancement of Machine Learning (ML) and Internet of Things (IoT) technologies, automated and intelligent solutions for crop disease detection have become feasible. The IoT-Based Crop Disease Recognition and Field Notification System proposes an intelligent system that combines machine learning–based image analysis with IoT-enabled monitoring to detect crop diseases at an early stage. The system uses an ESP32 microcontroller integrated with an ESP-CAM module to capture images of plant leaves. These images are analyzed using trained machine learning models to identify disease patterns and abnormalities. The detection results are communicated through an IoT platform, enabling remote monitoring and real-time alerts. An LCD display provides local status information, while a buzzer generates immediate alerts when a disease is detected. The system is designed to be cost-effective, scalable, and suitable for deployment in real agricultural environments. By enabling early disease identification and timely intervention, the proposed solution helps improve crop productivity, reduce losses, and promote smart and sustainable agricultural practices.

Plant phenotyping relevance

植物葉の画像を機械学習で解析し、病徴・異常を検出するシステムが研究の中心であり、植物病害状態の画像ベースフェノタイピングに該当する。

Code and data availability

The article describes an ESP32/ESP32-CAM ML-IoT crop disease detection prototype but provides no public dataset, image collection, code repository, trained model, or supplement with availability language. No paper-specific public assets are identified, and no allowed URLs are provided.

No evidence-backed public reproduction asset is currently recorded.