ABSTRACT
The application of remote sensing to the social sciences is an emerging research area. Recognizing that people’s behavior and values shape, and are shaped by, the environment in which they live, analysis of overhead imagery can characterize geographic factors related to economic status and levels of social connectivity in a region. Observables associated with economic well being include the presence of commercial infrastructure, house size, number and types of livestock, presence of vehicles, and access to transportation. The transportation and communication infrastructure also indicates the expected level of interactions among elements of the society. Other important factors may be inferred from indicators derived from the imagery, such as level and types of agricultural production, population density, access to improved roads, distances to schools and businesses, and attributes of communities. Using imagery data collected over sub-Saharan Africa, we present an initial exploration of the direct and indirect indicators derived from the imagery. We demonstrate a methodology for extracting relevant measures from the imagery, using a combination of human-guided and machine learning methods. Using two regions for comparison, we present an initial image-based characterization of the levels of urbanization. Plans for extending and validating these methods are discussed.
