Extracting road networks with the aid of log gabor filters

A new method is proposed to extract road networks in urban areas from high resolution satellite images. The first step is to segment road from other objects based on the real part of the transform of log Gabor function. An improved line segment matching method is then applied to the binary image to eliminate nonroad objects. After removing the short and irregular speckles in the segmented image, road segments may be detected easily. Connecting road segments finally produces the road networks. Experiments demonstrate that the proposed method can extract road networks in high-resolution satellite images efficiently.