Automatic Mapping of Physical Urban Problems Using Remotely Sensed Imagery

While big cities are expected to exercise cost-effective, evidence-based planning, many under reactive management, facing simultaneous problems and limited resources. This project develops a proof-of-concept workflow for the automatic monitoring of physical urban by combining remote sensing detection cartography visualization. The example problem treated was obstructive parking vehicles on pavements as proxy restricted mobility. Nine aerial images UK areas were processed deep learning object detector standard cars, achieving an F-score 70.72%. Two large scale map reports 200m wide produced, featuring car detections overlaps with topographic mapping features. Complementary analysis included calculation total window overlap per roadside pavement its change time. proposed method combines uniform city-wide coverage fast interpretation can inspire development professional planning tools.