数学形态学在SAR图像目标识别中的应用

The Application of Mathematical Morphology in Target Identification on SAR Image

提出了一种能够从复杂背景的SAR图像中识别出圆状目标的方法。该方法首先对SAR图像进行双参数恒虚警 (B -CFAR)预处理 ,排除明显的非目标区 ,增强下一步对目标的识别效果 ,然后着重分析了上下文相关加权的形态模板生成算法 ,在此基础上采用不同的数学形态学变换滤波方法对预处理后的SAR图像候选目标区进行识别 ,并与相关文献比较 ,对圆状目标得到了良好的识别效果。

In this paper, a method which can identify round targets from the complicated background of radar image is presented. First, this method performs pre-processing on the radar image using biparameter constant false alert rate (B-CFAR) which can wipe off the distinct non-target area and enhance the identified effect of the next step. Second, a context correlation method which can produces morphological template is analyzed. Last, we use different mathematical morphology to identify round target in the region of the pre-processing image. When using the method on the process of synthetic aperture radar (SAR) image, we obtain excellent effect in comparing with the correlative reference.