Crop Yield Prediction Using Satellite Images

Pavan Kumar M, Manav M, P C Sai Preetham, Dolanand S, Mahesh H.B

Department of Computer Science PES University Bangalore, India

Abstract—Crop yield estimation is a main task for economic and food management. The inadequate storage facilities and inadequate capital, and an unpredictable climate raise concerns for farmers and in the worst case, it may even lead to loss of life.

The yield prediction helps the farmers to improve crop produc- tion, calculate their savings, and to sanction loans according to the farmer’s capability directly from the office rather than going to a remote location for estimation. The increased availability of satellite data leads to further improvement. Deep learning helps the yield prediction applications using satellite images through which we can train the model and predict the future yield.

If there is a way to predict the yield by the end of the season, we can  make a decision  much easy. The project  implies that big businesses can use this model to optimize their price and inventory, the government can prepare for food shortages, and even a farmer be informed of appropriate selling prices if they know the regional yields.

Index Terms—agriculture, satellite crop yield prediction, ma- chine learning