A method for monitoring soil moisture using AI and a digital camera has been created

Engineers have improved the conventional digital camera using machine learning to control soil moisture and economically water. In addition, this one is more economical than analogs.

The United Nations (UN) predicts that many parts of the planet may not have enough fresh water to meet agricultural needs by 2050 if humankind continues with current resource use patterns.

One solution to this global dilemma is the development of more efficient irrigation, centered on accurate monitoring of soil moisture. It allows sensors to guide smart irrigation systems to ensure watering at the optimal time and speed.

The existing methods for measuring soil moisture are problematic: sensors located underground are sensitive to salts in the substrate and require special equipment to connect. In addition, thermal imaging cameras are expensive and depend on climatic conditions — the intensity of sunlight, fog, and clouds.

Researchers from the University of South Australia (UniSA) and the Technical University of Baghdad have developed a cost-effective alternative. It will make accurate soil monitoring easy and affordable in almost any circumstance.

A team of scientists, including UniSA engineers Dr. Ali Al-Naji and Professor Javan Chahl, have successfully tested the system. It uses a standard digital RGB camera to accurately monitor soil moisture over a wide range of environmental conditions.

“The system we have tested is simple, reliable and affordable, making it a promising technology to support precision farming,” says Dr. Al-Naji. “It is based on a standard video camera that analyzes differences in soil color to determine moisture content. We tested it at various distances, times and light levels and it was found to be very accurate. “

The camera was connected to an artificial neural network (ANN), a type of machine learning software that researchers had trained to recognize different levels of soil moisture in different sky conditions.

Using this INS, the monitoring system can potentially be trained to recognize specific soil conditions anywhere, allowing it to be customized for each user and updated according to changing climatic conditions for maximum accuracy.

Author: John Kessler
Graduated From the Massachusetts Institute of Technology. Previously, worked in various little-known media. Currently is an expert, editor and developer of Free News.
Function: Director
128 number 0.468330 time