A Feature to Believe In: Evaluating the Ability to use Contextual Features Derived from Multi-Scale Satellite Imagery to Map Spatial Patterns of Urban Attributes and Population Distributions
Open AccessWith an increasing global population, accurate and timely population counts are essential for urban planning and disaster management. Researchers have modeled population and socio-economic variables with contextual features derived from satellite imagery. Contextual features can be defined as the statistical quantification of edge patterns, pixel groups, gaps, textures, and the raw spectral signatures calculated over groups of pixels or neighborhoods. Previous research using contextual features has mainly used very-high spatial resolution imagery at subnational to city scales and has found strong correlations with population and poverty. This study evaluated the feasibility and accuracy of using contextual features derived from multi-scale satellite imagery to model elements of the human-modified landscape. Contextual features from very-high spatial resolution (< 2 m pixels) imagery and lower spatial resolution Sentinel-2 (10 m pixels) imagery in Sri Lanka and Belize were calculated, and those outputs were then correlated with OpenStreetMap building and road values. These relationships were compared to determine how spatial resolution impacts the predictive power and how different countries affect the relationship. The ability to predict the population density of the smallest census units available with Sentinel-2 contextual features was subsequently assessed. The results suggest that contextual features are able to map urban attributes well, with R-squared values ranging from 44% to 86%. Moreover, the degradation of spatial resolution does not significantly reduce the results, and for some urban attributes, the results actually improved. The findings also indicate that Sentinel-2 contextual features can explain up to 78% of the variation in population density in the study area. Finally, these contextual feature-based model results were compared with models built using nighttime lights. An analysis of the comparisons indicated that nighttime lights are less robust than contextual features in modeling urban attributes and population density.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.