扩展分形法结合B-CFAR在SAR图像目标检测中的应用

针对文献1]的扩展分形(Extended Fractal．EF)方法，分析了EF特征的优劣性。结合SAR图像的统计特性，本文将EF检测与双参数恒虚警车(Bi-parameter Constant False Alarm Rate B-CFAR)检测相结合．提出两个改进方案：一、利用两种特征的融合信息在决策层决策的检测方案；二、先用B-CFAR增强感兴趣目标区域(Region of Interest．ROI)．在此处理之后再用EF法检测。我们将这两种方案运用在复杂的二维SAR图像飞机检测中．得到了良好的检测效果。

Application of Extended Fractal with B-CFAR to Target Detection on SAR Image

Two improved methods combining Extended Fractal (EF) detection presented by reference~(1]) with Bi-parameter Constant False Alarm Rate (B-CFAR) detection are given in this paper by analyzing the advantages and disadvantages of EF features and combining synthetic aperture radar image's statistic characters. One is double-feature fusion technique~(2]) used in decision-making. The other is to use EF detection on regions of interest that are firstly enhanced by pre-processing with B-CFAR detection. Excellent detection effect can be obtained by using the two improved methods for airplane detection on the complicated two-dimension SAR image.