Smart Irrigation IoT Solution using Transfer Learning for Neural Networks

TL;DR

This paper presents a smart greenhouse irrigation system leveraging neural networks and IoT, utilizing transfer learning to enhance efficiency, speed, and integration of climate sensors, demonstrating superior performance over traditional methods.

Contribution

Introduces a transfer learning approach to neural networks for IoT-based smart irrigation, reducing processing power needs and enabling climate sensor integration.

Findings

Neural networks outperform support vector regression in moisture prediction.

Transfer learning accelerates training with limited data.

Proposed IoT architecture provides a comprehensive smart irrigation solution.

Abstract

In this paper we develop a reliable system for smart irrigation of greenhouses using artificial neural networks, and an IoT architecture. Our solution uses four sensors in different layers of soil to predict future moisture. Using a dataset we collected by running experiments on different soils, we show high performance of neural networks compared to existing alternative method of support vector regression. To reduce the processing power of neural network for the IoT edge devices, we propose using transfer learning. Transfer learning also speeds up training performance with small amount of training data, and allows integrating climate sensors to a pre-trained model, which are the other two challenges of smart irrigation of greenhouses. Our proposed IoT architecture shows a complete solution for smart irrigation.