Semi-automatic Extraction of Ribbon Road from High Resolution Remotely Sensed Imagery by a T-shaped Template Matching

Based on the characteristics of roads on high-resolution remotely sensed imagery,a T-shaped template matching is proposed to track ribbon roads with much vehicles occlusion and shadow of colonnade on roads' edges.The T-shaped template matching integrates the merits of profile matching and template matching.Our tracker uses angular texture signature to locate the starting point and to measure the road width and road direction,and employs a T-shaped template matching to predict the sequential road points.It is proved a more robust semi-automatic road tracker.