The traveling public judges the quality of a road mostly by its roughness and/or ride quality. Hence, mapping, monitoring, and maintaining adequate pavement smoothness is of high importance to State Departments of Transportation in the US. Current methods rely mostly on in situ measurements and are, therefore, time consuming and costly when applied at the network scale. This paper studies the applicability of satellite radar remote sensing data, specifically, high-resolution Synthetic Aperture Radar (SAR) data acquired at X-band, to the network-wide mapping of pavement roughness of roads in the US. Based on a comparison of high-resolution X-band Cosmo-SkyMed images with road roughness data in the form of International Roughness Index (IRI) measurements, we found that X-band radar brightness generally increases when pavement roughness worsens. Based on these findings, we developed and inverted a model to distinguish well maintained road segments from segments in need of repair. Over test sites in Augusta County, VA, we found that our classification scheme reaches an overall accuracy of 92.6%. This study illustrates the capacity of X-band SAR for pavement roughness mapping and suggests that incorporating SAR into DOT operations could be beneficial.
DCN-Based Spatial Features for Improving Parcel-Based Crop Classification Using High-Resolution Optical Images and Multi-Temporal SAR Data