论文部分内容阅读
A method used for recognition and understanding of airfield based on mathematical morphology is proposed in this paper. The new approach can be divided into three steps. First, to extract the typical geometric structure features of airfield, a segmentation method called recursive Otsu algorithm is employed on an airfield image. Second, thinning and shrinking algorithms are utilized to obtain the contour of airfield with single pixel and to remove diffused small particles. Finally, Radon transform is adopted to extract two typical and important components, primary and secondary runways of airfield exactly. At the same time, region growing algorithm is exploited to get the other components such as parking apron and garages. The experimental results demonstrate that the proposed method gives good performance.
A method used for recognition and understanding of airfield based on mathematical morphology is proposed in this paper. First, to extract the typical geometric structure features of airfield, a segmentation method called recursive Otsu algorithm is employed Second, thinning and shrinking algorithms are utilized to obtain the contour of airfield with single pixel and to remove diffused small particles. Finally, Radon transform is adopted to extract two typical and important components, primary and secondary runways of air exactly At the same time, region growing algorithm is exploited to get the other components such as parking apron and garages. The experimental results demonstrate that the proposed method gives good performance.