Harnessing Satellite Imagery with Convolutional Neural Networks for Poverty Prediction

Investigating Poverty Prediction from Space looks at applying Convolutional Neural Networks (CNNs) to satellite imagery in order to learn more about how poverty is distributed. Efficient approaches are required to map poverty, since over a billion people live below the worldwide poverty line. Conventional census-based methods are expensive and time-consuming. This study shows how well CNNs can forecast poverty levels by utilising the wealth of high-resolution satellite data. Focusing on Mali, Ethiopia, Malawi, and Nigeria, our CNN models, achieved promising accuracies, with the custom CNN model achieving 90% accuracy. This innovative approach aids in resource allocation and policy formulation to fight against poverty effectively.