Soil Moisture and Heat Level Prediction for Plant Health Monitoring Using Deep Learning with Gannet Namib Beetle...

Area of Science:

• Agricultural Science
• Computer Science
• Environmental Science

Background:

• Effective plant health monitoring is vital for global food security.
• Soil moisture and temperature are critical factors influencing plant growth and yield.
• Predictive insights into soil conditions enable proactive disease management and yield maximization.

Purpose of the Study:

• To develop an Internet of Things (IoT) based system for real-time plant health monitoring.
• To forecast soil moisture and temperature levels for early disease detection and yield optimization.
• To enhance data transmission efficiency and prediction accuracy in agricultural monitoring systems.

Main Methods:

• Utilized an Internet of Things (IoT) environment for collecting soil data.
• Implemented a cluster head (CH) selection and routing technique using Gannet Namib beetle optimization (GNBO) for data transmission.
• Employed Convolutional long short term memory (Conv-LSTM) for predicting soil moisture and heat levels, with hyperparameters optimized by GNBO.

Main Results:

• The GNBO-Conv-LSTM model demonstrated efficiency in forecasting soil conditions.
• Key performance metrics included a link lifetime (LLT) of 0.675, energy consumption of 0.478 J, and a delay of 0.092 ms.
• The model achieved a positive predictive value (PPV) of 0.882 and a true negative rate (TNR) of 0.875.

Conclusions:

• The proposed GNBO-optimized Conv-LSTM model offers a robust solution for plant health monitoring.
• Accurate soil condition prediction through this IoT system can significantly aid farmers in mitigating crop diseases.
• This approach contributes to sustainable agriculture by maximizing crop yield and ensuring food supply stability.