Spiking Neural Network (SNN) for Crop Yield Prediction

Abstract

Crop yield prediction focuses mostly on agricultural research, which have an enormous impact on taking decisions for example import-export, price, along with crop management. Accurate forecasting with well-timed projections is critical, but it is a challenging undertaking owing to various complicated aspects. There are few examples of crops that can be utilized to forecast crop yields like Wheat, peas, rice, pulses, tea, sugar cane, green houses, cotton, soybeans, and corn. Agriculture needs massive datasets and awareness practices. Meteorological conditions, components of soil, management methods, genotype, and their connections are utilized to predict corn yield. Optimal crop growth frequently requires a detailed knowledge of the operational relationships among yield and these interaction parameters, that needs large datasets and difficult algorithms to demonstrate. Several Machine Learning models, Deep Learning models, and Artificial Neural Network methods are used to forecast. Convolutional Neural Networks (CNN), Spiking Neural Networks (SNN), and Recurrent Neural Networks (RNN) are used to estimate corn production (RNN). By integrating RNN and SNN models, each model functioning was improved.