Economic well-being assessment: a review of traditional and remote sensing approaches

This paper reviews the evolution of economic well-being assessment, examining traditional methods based on surveys and statistics alongside the emergent field of satellite remote sensing. Traditional approaches, employing indicators like GDP, HDI, and MPI, offer established methodologies but face limitations in data accessibility, spatial coverage, and capturing dynamic changes. Satellite remote sensing, utilizing nighttime light, daytime imagery, and derived data like NDVI, overcomes these constraints by providing large-scale, timely, and periodic surface information. Furthermore, we analyze methods for large-scale economic well-being assessment using remote sensing, encompassing statistical analysis, machine learning, deep learning, and transfer learning. Finally, we explore future directions, emphasizing the development of more comprehensive indicators, multi-source data fusion integrating subjective well-being, and advancements in deep neural networks to improve accuracy, generalization, and interpretability for robust, large-scale economic well-being assessment and poverty reduction strategies.