Abstract
AbstractAgriculture is the main part of India’s economy which provides food security for the country and produces several raw materials for the industries. The development in agriculture is an important aspect in the nearby future. Sugarcane crop is one of the highest producing crops in India, and Maharashtra state is the second highest producer of the sugarcane. In this paper, a novel approach for the yield prediction of the sugarcane crop is proposed based on the weather and soil parameters, normalized difference vegetation index (NDVI), and several machine learning regression techniques. The model is verified using historical data set for the sugarcane crop. The model consists of three stages—(I) Prediction of the weather parameters, (II) prediction of NDVI using weather parameters as input, (III) yield prediction using stage I and II as input. The decision tree regressor gives the highest accuracy of 91.5% for the final model of sugarcane crop yield prediction.KeywordsMachine learningCrop yield productionRemote sensing dataNormalized difference vegetation index (NDVI)Time series dataYield predictionWeather data
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have