Enhanced reptile search optimization with convolutional autoencoder for soil nutrient classification model

Area of Science:

• Agricultural Science
• Environmental Science
• Computer Science

Background:

• Soil nutrients are crucial for soil fertility and crop productivity.
• Traditional soil testing methods are time-consuming and impractical for real-time analysis.
• Accurate prediction of soil nutrients and pH is essential for optimizing agricultural practices.

Purpose of the Study:

• To develop an efficient machine learning model for predicting soil nutrient levels and pH.
• To enhance soil fertility assessment using advanced computational techniques.
• To improve crop productivity and reduce agricultural input costs.

Main Methods:

• Development of an enhanced reptile search optimization with convolutional autoencoder (ERSOCAE-SNC) model.
• Utilizing the convolutional autoencoder (CAE) for classifying nutrient levels (Phosphorus, Potassium, Organic Carbon, Boron) and soil pH.
• Employing the ERSO algorithm for hyperparameter tuning to improve CAE classification performance.

Main Results:

• The ERSOCAE-SNC model achieved high accuracy: 98.99% for soil nutrients and 99.12% for soil pH.
• The model outperformed existing approaches in soil fertility prediction.
• Simulations confirmed the model's effectiveness and efficiency.

Conclusions:

• The ERSOCAE-SNC model offers a practical solution for real-time soil nutrient and pH prediction.
• Implementation can aid government initiatives in managing soil nutrient deficiency and improving environmental quality.
• The model promises reduced fertilizer expenditures and time savings for soil experts.