CNNs for analyzing urban visual experience
Urban Spatial Openness Index (USOI), captures the spatial openness of open spaces in urban environments. Open spaces are defined as non-built areas: streets, squares parks, rivers. USOI values range from 0% (no visible space at all) to 100% (complete spatial openness), and work best as a comparative tool between alternatives. Higher USOI values have been associated with lower perceived density.
USOI results are visualized as heatmaps, where each cell is colored according to its respective USOI value, at a spatial resolution of 5-by-5 meters. The trained model takes height image and predicts respective USOI values and can estimate city-wide spatial openness in seconds. Results account for the effects of buildings’ locations and heights on the spatial openness around them. Trees and topography are not considered in this model; therefore, it is more suitable for flat places.
Keywords
Visibility Analysis
Perceived Density
Machine Learning
CNN