A new method of road extraction from high-resolution remote sensing imagery

It is still an open problem to extract object features from high-resolution remote sensing images, although this topic has been intensively investigated and many methods have been carried out. This thesis focuses on modern urban roads in the following four steps, namely imagery pre-processing, threshold calculation, feature extraction for straight line and curved line and target reconstruction. From this perspective, a new and semi-automatic approach is proposed based on the phase classification. Firstly, the basic road network can be obtained from high-resolution remote sensing images based on grey level mathematical morphology and canny algorithm. Secondly, the road information can be accurately extracted by means of the "grey" parameters, which are various for different kinds of road models according to the theory of phase-based classification. Thirdly, the proposed method can also be employed to elevate urban highways, especially for their curve parts. The experimental results demonstrate that the proposed extraction method can obtain a reasonable result.