大幅场景高分辨率SAR图像机场自动检测

针对现有大场景高分辨率SAR图像机场检测方法中所采用的图像分割方法效果不佳、适应性不强,直线特征提取方法耗时长且易产生断裂的问题,提出一种在大幅场景SAR图像中自动检测机场的方法。首先,提出基于PPB滤波的模糊C均值算法进行图像分割,利用区域特征筛选出感兴趣区域(ROI);然后,采用启发式搜索快速有效地提取线基元,并提出新的共线性测度方法减少线基元连接中的误连接和欠连接;最后,利用平行线特征和灰度特征实现机场的检测。实验结果验证了该方法的有效性。

Airport Automatic Detection in Large-Scale SAR Image with High Resolution

The adopted image segmentation methods in airport detection from large-scale SAR image with high reso-lution are poor in performing, and the adopted line extraction methods are time-consuming and prone to fracture. Aiming at these problems, an algorithm of airport automatic detection in large-scale SAR image with high resolution was proposed. Firstly, the fuzzy c-means based on PPB filter algorithm is put forward for image segmentation, and the regions of interest are screened by region features. Secondly, linear primitives are extracted by heuristic search fastly and efficiently and a new method for collinearity measuring is proposed to reduce connection error and insuffi-cient. Finally, airports are detected by the feature of parallel and gray level. The experimental results verified the effectiveness of the method in this paper.