大场景SAR图像中提取小型港口区域技术研究

发布时间：2018-05-07 00:38

本文选题：SAR图像 + 小型港口提取　；参考：《中国图象图形学报》2017年08期

【摘要】：目的漫长的海岸线上有很多由简单码头和突堤组成的小型港口,它们不像大型港口那样时刻被监视着,但其作为舰船的临时停靠场所,也是需要重点关注的区域,有必要对其区域进行自动检测。方法本文在充分分析小型港口特征的基础之上,根据其海岸线轮廓在不同尺度下均能表现出丰富角点的特点,构造了提取小型港口区域的完整流程,包括采用多尺度角点检测提取港口潜在区域、采用新型突堤检测方法精提取港口区域、采用改进的岸线封闭性测度法鉴别去除虚假港口3个环节,通过实际SAR图像仿真验证了方法的优越性。结果利用本文所提出的方法,处理两幅典型的港口区域SAR图像:Radarsat-2烟台港1 m分辨率的SAR图像和Terra SAR-X印度维沙卡帕特南港1 m分辨率的SAR图像,并与文献4中所述的方法进行对比,发现经过本文方法处理之后,虚警率从10%降到了6.6%,准确率从91.9%提高到了93.3%,但是由于计算流程较复杂,导致处理时间从11.58 s增加到了13.26 s。结论本文针对小型港口的特点,提出了港口检测的完整的流程。实验结果表明,该方法的虚警率更低、准确性更高,但是存在运算速度慢的缺点,这是下一步需要优化的地方。该方法适用于大场景SAR图像中快速准确地检测出小型港口区域,可用于监视那些由简单码头和突堤组成的舰船临时停靠场所。
[Abstract]:There are many small ports of simple wharves and jetty along the long coastline that are not always monitored like large ports, but they are places of temporary docking for ships and areas that need to be focused on. It is necessary to detect its area automatically. Methods on the basis of fully analyzing the characteristics of small ports and according to the characteristic that the coastline contour can show rich corner points at different scales, the complete process of extracting small port area is constructed in this paper. Including using multi-scale corner detection to extract the potential port area, using a new method of breakwater detection to extract the port area, and using the improved shoreline closure measure method to identify and remove the false port three links. The superiority of the method is verified by actual SAR image simulation. Results using the method proposed in this paper, two typical SAR images of the port area: 1 m SAR image of Yantai port and 1 m resolution SAR image of Terra SAR-X port of Vishaka Patnam, India, were processed. Compared with the method mentioned in reference 4, it is found that the false alarm rate is reduced from 10% to 6.6%, the accuracy rate is increased from 91.9% to 93.3s, but the processing time is increased from 11.58 s to 13.26 s due to the complexity of the calculation process. Conclusion according to the characteristics of small ports, a complete flow of port detection is put forward in this paper. The experimental results show that the false alarm rate is lower and the accuracy is higher, but there is a disadvantage of slow operation speed, which is the next step to be optimized. This method is suitable for fast and accurate detection of small port areas in large scene SAR images and can be used to monitor temporary docking sites of ships consisting of simple wharves and breakwaters.
【作者单位】：海军航空工程学院科研部;海军航空工程学院电子信息工程系;海军航空工程学院信息融合研究所;
【基金】：国家自然科学基金项目(91538201)~~
【分类号】：TN957.52

【相似文献】