A Framework for Accurate Food Crop Yield Prediction from Multispectral Imagery and Meteorological Data

Food insecurity occurs due to the impact of climate change and intense global conditions. Thus, understanding the crop farming plans and monitoring crop yields have become major tasks for decision makers. Previous work has applied remote sensing techniques and empirical methods to predict the yields and analyse the relationships between spectral indices and historical crop yield data. However, a limitation of these studies is that they do not extract the values of spectral indices by crop types, which can cause inaccurate results when investigating the correlations between the yield and the spectral indices. This research develops a yield prediction framework with historical crop maps by unsupervised classification with zero ground truth using Sentinel-2 imagery to retrieve the values of spectral indices of winter barley. The extracted spectral indices, the meteorological and historical yield data in North Norfolk, UK are implemented in 1D Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) and CNN-LSTM for winter barley yield predictions. LSTM with Soil-Adjusted Vegetation Index (SAVI) has outstanding performance overall and the best result approaches a Mean Square Error (MSE) of 0.21 kg/hectare and a Mean Absolute Error (MAE) of 13.63 kg/hectare. SAVI is the best predictor due to the strong positive correlation with the yield of winter barley. The developed framework with unsupervised crop classification and LSTM-SAVI can be applied on multiple crop types and in different regions using opensource datasets, the historical yields, the spectral indices and the meteorological data. Correlations between these datasets indicate that higher maximum and minimum SAVI and temperature and sun hours at the germination and seedling growth stages increase the yields of winter barley but excess rainfall with higher NDMI at the tillering stage and sun hours at the stem elongation, flowering and grain filling stages lead to a decline in the yields of winter barley.