In this paper, a road extracton method is proposed by combining kernel-based Fisher linear discriminant (FLD) classification and shape feature. This method has four main steps: First, the color information of labeled samples is extracted. Then kernel-based Fisher linear discriminant is used to implement feature classification to segment the images into two categories: road and non-road, according to the information extracted before. After that, the road shape optimization features are used to remove erroneous extraction. Finally, morphological processing are used to optimize the extraction results.Experiment results show that the proposed method in this paper can realize the extraction of road from remote sensing image with color information.
