基于异常检测与双层筛选机制的SAR图像舰船检测方法

发布时间：2019-04-09 07:27

【摘要】：针对现有合成孔径雷达(SAR)图像舰船目标智能检测算法中筛选误差较大的问题,提出一种新的SAR图像舰船目标检测方法。该方法将高光谱图像异常检测理论引入到SAR图像舰船目标检测处理中。通过图像转换将SAR图像转换成高光谱类型图像,采用异常检测算法实现舰船目标的检测预处理,得到感兴趣区域二值图。运用双层筛选机制,实现背景杂波的准确建模和舰船目标的快速检测。实验结果表明,该算法能够降低筛选误差,有效地消除虚假目标和旁瓣干扰,具有更好的结构保真度。
[Abstract]:A new method of ship target detection based on synthetic Aperture Radar (SAR) image is proposed to solve the problem of large filtering error in the existing algorithms of ship target intelligent detection in synthetic aperture radar (SAR) images. In this method, the theory of hyperspectral anomaly detection is introduced into the ship target detection of SAR image. The SAR image is converted into hyperspectral image by image conversion, and the detection preprocessing of ship target is realized by using anomaly detection algorithm, and the binary map of the region of interest is obtained. The background clutter modeling and the fast detection of ship targets are realized by using the double-layer filtering mechanism. The experimental results show that the proposed algorithm can reduce the filter error, effectively eliminate false targets and sidelobe interference, and has better structure fidelity.
【作者单位】：国防科学技术大学电子科学与工程学院;
【基金】：国家自然科学基金(61171135)
【分类号】：TN957.52

【相似文献】