An Integrated Real Time IoT System for Soil Monitoring and Deep Learning Based Plant Health Assessment
Year: 2026
Type: article
Abstract: -Agriculture plays an important role in economic development and food production.Effective agricultural monitoring requires information about soil conditions andplant health. Conventional monitoring practices may not provide integratedinformation about soil moisture, nutrient levels, and visible plant health. Advancedtechnologies such as the Internet of Things (IoT) and deep learning (DL) cansupport intelligent agricultural monitoring. In this paper an integrated IoT and DLbased framework is proposed for agricultural soil monitoring and plant leaf healthassessment. Here a capacitive soil moisture sensor and RS485 based NPK sensorare interfaced with an ESP32 microcontroller to measure soil moisture andnitrogen, phosphorus, and potassium concentrations. The acquired measurementsare transmitted to a Firebase cloud database via Wi-Fi. Finally, it displayed on aweb-based dashboard for real time monitoring of soil conditions. In parallel, theproposed Chry-CDNet model analyses chrysanthemum leaf images. It classifiesplant health into two categories: healthy and diseased. The model achievedclassification accuracy of:96.67±0.11%. This result demonstrates its effectivenessin assessing plant health using leaf images. The framework combines sensorderived soil information with visual plant assessment to provide a comprehensiveview of chrysanthemum cultivation conditions. The system supports data drivencrop monitoring and promotes precision floriculture.
Author: Sarkar, Moumita
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7233546552
FWCI:
Cites:
Cited by:
Related to:
