Runways detection based on scattering similarity and structural characteristics

In this paper, a new algorithm of runways detection based on classification is proposed for PolSAR image, in which scattering similarity and structural characteristics of airport runways are used as target feature for classification. Firstly, training sample templates are built by initial terrain classification with scattering similarity. Then, Bayesian classification is used to classify the whole image. Finally, morphology filtering and the runways structural features such as parallels, topologic property are used to detect real runways. Multi-look fully polarimetric SAR data acquired by U.S. UAVSAR systems is used to verify the proposed algorithm. Experimental results show that the novel method can detect runways effectively and has a low false alarm rate. Besides, the detected runways keep an intact structure and clear outlines.