Open Access

ARTICLE

Design and Implementation of an IoT-Based Irrigation System with Surveillance Camera

Moina-yndi Ibrahim¹, Warda Soulaimana², Zhenjie Zhao^3,*

1 School of Computer Science, Nanjing University of Information Science and Technology, Nanjing, China
2 Department of MIAGE-ACSI, Institute of Digital Sciences, Management Cognition, Nancy, France
3 College of Artificial Intelligence, Nankai University, Tianjin, China

* Corresponding Author: Zhenjie Zhao. Email:

Journal on Internet of Things 2026, 8, 67-86. https://doi.org/10.32604/jiot.2026.078735

Received 07 January 2026; Accepted 03 March 2026; Issue published 07 April 2026

Abstract

The increasing demand for efficient agricultural water management, exacerbated by population growth and climate change, has spurred the development of Internet of Things (IoT)-based smart irrigation systems as an alternative to inefficient traditional methods that waste water and reduce crop productivity. This paper presents a low-cost IoT irrigation platform that integrates real-time camera surveillance for enhanced farm monitoring and precise water management. The system employs an Espressif System (ESP8266) microcontroller to automate irrigation control based on soil moisture readings and an Espressif System ESP32 Camera Module (ESP32-CAM) to provide live imaging of the agricultural plot, with both devices connected to cloud services via Wireless Fidelity (Wi-Fi). System architecture was modeled using Unified Modeling Language (UML) diagrams, including use-case and sequence diagrams, to ensure clarity and robust design. The IoT framework follows a four-layer model encompassing field perception, network transmission, cloud-based processing via the Blynk platform, and a mobile/web application layer for remote user interaction. A smartphone dashboard allows farmers to visualize real-time sensor data, monitor live video feeds, and control irrigation components remotely. Experimental evaluation over a 14-day outdoor deployment involving 47 automated irrigation events demonstrated a mean pump activation response time of 1.2 ± 0.21 s, 100% pump activation accuracy across 30 controlled trials, and system uptime of 96.4%. A baseline comparison showed that camera integration reduced fault detection latency approximately six-fold relative to a sensor-only configuration (15 vs. 90 s). The key scientific contributions include: (i) the integration of a ground-level ESP32-CAM into an automated threshold-triggered irrigation loop with empirical fault-detection validation; (ii) quantitative benchmarking against a sensor-only baseline; and (iii) formal UML-based system specification applied to an IoT irrigation context. The proposed platform offers a practical, scalable smart irrigation solution unifying automated control, cloud processing, and live visual monitoring within a single framework.

---

Keywords

---