Straight Roads Extraction by Fusion of Spoke Wheel and Region Growing

Semi-automatic methods are regarded as more practical than automatic methods to obtain and/or update the vector road data. But road tracking methods lose their efficiencies when they are applied to track a long straight roads step by step compared to manual digitalizing. So an efficient semi-automatic method is proposed in this paper. Particularly, spoke wheel algorithm and region growing algorithm are combined together to extract a rectangle corresponding to a short road segment. That above process is performed at each pixel along a line segment between two seed points, which derives a number of rectangles. Then statistical analysis of histograms of directions and widths of the obtained rectangles is made to determine an optimal rectangle. At last, a precise road model is determined by the above optimal rectangle. Extensive experiments show that our method is capable of efficiently extracting a straight long road with a high spatial accuracy, even various types of noises existing. * Corresponding author. [email protected]