Related Experiment Video
Updated: Aug 16, 2026
08:20
In Situ Soil Moisture Sensors in Undisturbed Soils
Published on: November 18, 2022
A hybrid framework combining microwave sensing and deep learning for real-time soil volumetric water content
Adel M Soliman1, Amr Elbrashy2, Amr H Hussien3
1Department of Electronics and Communications, Faculty of Engineering, Horus University-Egypt (HUE), New Damietta, 34517, Egypt.
Scientific Reports
|August 14, 2026
Summary
This study presents a real-time soil moisture measurement system using microwave sensing and machine learning. It provides fast, accurate volumetric water content (VWC) estimates, improving geotechnical decisions and reducing delays from traditional methods.
Area of Science:
- Geotechnical Engineering
- Remote Sensing
- Machine Learning
Background:
- Traditional soil moisture measurement (gravimetric) is destructive, time-consuming (24-hour delay), and energy-intensive.
- Accurate, real-time soil moisture data is crucial for geotechnical engineering and agriculture.
Purpose of the Study:
- To develop and validate an integrated, non-destructive, real-time soil moisture measurement system.
- To overcome the limitations of conventional soil moisture assessment methods.
Main Methods:
- Utilized a 2.4 GHz microwave/RF sensing system integrated with edge computing (Raspberry Pi 5).
- Employed a convolutional neural network (CNN) for soil texture classification (sand, clay, mixed).
- Applied polynomial regression models for precise prediction of volumetric water content (VWC).
Main Results:
- Achieved high agreement between measured and predicted VWC across various soil textures.
- Demonstrated exceptional prediction accuracies: 99.8% (sand), 99.7% (clay), and 99.4% (mixed).
- Reduced measurement time from 24 hours to real-time.
Conclusions:
- The developed system offers a reliable, cost-effective, and scalable solution for continuous soil monitoring.
- Enables rapid, on-site VWC estimation for urgent geotechnical decisions and proactive risk management.
- Significant potential for applications in civil engineering and precision agriculture.