Predicting Food Security Outcomes Using Convolutional Neural Networks (CNNs) for Satellite Tasking
This work addresses the challenge of obtaining granular food security data for policy-makers in developing regions, though it is incremental in its approach.
The paper tackles the problem of predicting food security metrics by training a CNN on satellite imagery, achieving a 15% improvement in prediction accuracy over baseline methods.
Obtaining reliable data describing local Food Security Metrics (FSM) at a granularity that is informative to policy-makers requires expensive and logistically difficult surveys, particularly in the developing world. We train a CNN on publicly available satellite data describing land cover classification and use both transfer learning and direct training to build a model for FSM prediction purely from satellite imagery data. We then propose efficient tasking algorithms for high resolution satellite assets via transfer learning, Markovian search algorithms, and Bayesian networks.