K分布下极化SAR图像CFAR检测新方法

在极化合成孔径雷达（PolSAR）图像杂波服从K分布下，已有的基于多视极化白化滤波器（MPWF）的恒虚警（CFAR）检测方法不适用于等效视数为非整数情形.针对此问题，提出了一种具有解析虚警概率表达式的CFAR检测方法.在乘积模型下引入Gamma分布纹理变量后，对MPWF检测量的概率密度函数（PDF）积分得到其累积分布函数（CDF），进而得到虚警概率关于检测门限的解析形式.仿真实验结果表明新方法良好的虚警拟合效果，此外在目标杂波比（TCR）较大时，新方法检测概率更高，反之，检测概率低于已有方法.需说明的是，目前K分布下多视极化匹配滤波（MPMF）检测方法的虚警概率表达式存在错误，因此对K分布下的MPMF检测方法进行了修正并推导出了正确的虚警表达式.

A Novel CFAR Detection Algorithm in K-Distributed Clutter for Polarimetric SAR Imagery

In the case that the clutter of the polarimetric synthetic aperture radar (PolSAR) imagery follows K distribution,the existing constant false alarm rate (CFAR) detection algorithms based on multi-look polarization whitening filter (MPWF) cannot be applied to the condition that the equivalent number of looks is a non-integer.To solve the problem,a CFAR detection algorithm with analytical false alarm probability expression is proposed.After introducing the Gamma distribution texture under the product model,the probability distribution function (PDF) of the MPWF metric is integrated to obtain its cumulative distribution function (CDF),and then the analytical form of the false alarm probability with respect to the detection threshold is obtained.The simulation results show that the new algorithm has a good false alarm fitting effect.In addition,when the target clutter ratio (TCR) is large,the new algorithm has a higher detection probability.Otherwise,the detection probability is lower than the existing algorithm.It is worth pointing out that the current false alarm probability expression of multi-look polarization matched filter (MPMF) under K distribution is wrong,so the current detection algorithm based on MPMF detection metric is modified and the correct expression of the false alarm probability is derived.