Enhanced Soil Nutrient-NPK Measurement Using Electrical Conductivity and Temperature

To satisfy the growing demand of population, opti-mizing the level of soil micronutrients, i.e., Nitrogen (N), Phosphorus (P), and Potassium (K) (known as NPK), plays a crucial role. However, the traditional methods for determining the soil nutrient level are accurate but time-consuming, expensive, and laborious. This study uses a deep learning approach to predict the soil NPK levels, holding electrical conductivity (EC) and temperature (T) as inputs and using a 7-in-1 soil sensor. A dataset was developed using different soil samples. A neural network model was developed to train and test the discovered dataset. The model achieved promising results with 99% accuracy with lower mean squared error (MSE) value, demonstrating rapid and accurate soil nutrient sensing potential. This research provides a cost-effective solution to model a soil sensor to streamline precision agriculture and optimize fertilization techniques.