CROPSENTINEL USING MACHINE LEARNING AND IOT

Abstract

Efficient water management is a critical challenge in modern agriculture, where traditional irrigation methods often lead to water wastage and reduced crop productivity. To address this issue, this paper presents an IoT-based smart irrigation system that integrates real-time image acquisition with machine learning–based decision making. The proposed system uses an ESP32-CAM module to periodically capture images of crop leaves and transmit them wirelessly to a processing unit for analysis. A machine learning model trained on labeled crop images evaluates the visual condition of the leaves to determine plant stress and irrigation requirements. By analyzing features such as leaf dryness and visible health indicators, the system makes intelligent irrigation decisions without relying solely on manual observation or fixed schedules. This approach enables more accurate and adaptive water usage compared to conventional irrigation systems. The system architecture is designed to be low-cost, scalable, and energy-efficient, making it suitable for small and medium-scale agricultural applications. Hardware implementation includes the ESP32-CAM module with integrated Wi-Fi, while the software framework utilizes image preprocessing and deep learning techniques for classification. Experimental results show that the system can successfully capture and analyze crop images and provide timely irrigation decisions. This work demonstrates the effectiveness of combining IoT and machine learning technologies for precision agriculture and sustainable water resource management.

Abstract from the paper’s OpenAlex record; open copy licensed cc-by-nc. Read at the source ↗

The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.