论文部分内容阅读
机场在军事和交通运输领域都有很重要的作用,对它的自动检测具有重大意义。该文提出了一种利用全极化合成孔径雷达(PolSAR)图像检测机场和跑道的方法。首先,通过一种改进的Freeman分解提取3种散射机制的能量作为像素点的特征,利用模糊C-均值聚类(FCM)算法分割图像来获得疑似机场目标区域(region of interest,ROI)。然后采用基于复wishart分布的K-均值聚类算法精细分割ROI,得到完整的机场跑道结构;提出一种检测机场主跑道的方法进行ROI辨识,最终确定机场目标。采用多组PolSAR数据进行验证,并比较了单、双和全极化图像的机场检测效果。结果表明:该方法具有较高的检测率和较低的虚警率,并且具有较好的抗噪性能。
The airport plays a very important role in the military and transportation fields and has great significance for its automatic detection. This paper presents a method of detecting airports and runways using fully polarized synthetic aperture radar (PolSAR) images. Firstly, an improved Freeman decomposition was used to extract the energy of the three scattering mechanisms as the feature of the pixel. The image of the airport was segmented by fuzzy C-means clustering (FCM) to obtain the region of interest (ROI) of the suspected airport. Then the ROI is finely partitioned by K-means clustering algorithm based on complex wishart distribution to get the complete structure of airport runway. A method of detecting the main runway of the airport is proposed to identify the ROI and finally determine the airport target. Multiple sets of PolSAR data were used for verification, and the airport detection effects of single, double and full polarizations were compared. The results show that this method has higher detection rate and lower false alarm rate, and has better anti-noise performance.