论文部分内容阅读
为了对极化合成孔径雷达(polSAR)图像中小型港口目标进行自动检测,在分析小型港口特性的基础上,提出了一种基于岸线特征点合并的检测方法。首先,使用极化SAR图像水平集分割算法实现精确的海岸线提取,并通过数字曲线分裂归并算法提取海岸线轮廓特征点;然后针对小型港口轮廓特征点比非港口区域轮廓的密集的特性,提出了一种岸线特征点合并算法实现港口检测。分别用RADARSAT-2系统获取的新加坡和湛江海岸区域极化SAR数据对提出方法进行了试验。实验结果表明,该方法能够正确地检测沿岸小型港口。
In order to automatically detect small and medium-sized ports in the polsar image, a new detection method based on the combination of shoreline feature points is proposed based on the analysis of the characteristics of small ports. First, we use the SAR image level set segmentation algorithm to achieve the accurate coastline extraction, and extract the coastline feature points by the digital curve splitting and merging algorithm. Then, according to the dense characteristic of small port contour feature points than the non-port area contour, Kinds of Coastline Feature Points Merging Algorithm for Port Detection. The proposed method was tested using the regional SAR data from the Singapore and Zhanjiang coasts acquired by the RADARSAT-2 system. Experimental results show that this method can detect small ports along the coast correctly.