Beta分布下基于白化滤波的极化SAR图像海面舰船目标CFAR检测方法

A CFAR Detection Method of Ship Targets in Polarimetric SAR Imagery Based on Whitening Filter Under Beta Distributed Texture

在杂波纹理服从Beta分布的极化合成孔径雷达(Polarimetric Synthetic Aperture Radar,POLSAR)图像目标检测中,提出了一种基于多视极化白化滤波(Multilook Polarimetric Whitening Filter,MPWF)的恒虚警(Constant False Alarm Rate,CFAR)检测解析新方法.首先,假设乘积模型中纹理变量服从Beta分布,推导得到MPWF检测量的概率密度函数(Probability Density Function,PDF).然后,对概率密度函数积分得到虚警概率关于检测门限的解析式,并设计相应的CFAR检测流程.最后,提出了基于MPWF的对数累积量估计方法,对Beta分布纹理变量参数u和v进行估计.通过实测数据验证了新方法的有效性.实验结果表明Beta分布对某些区域的极化SAR数据有更好的拟合效果,同时新方法与已有方法相比具有更好的CFAR保持能力.

A new constant false alarm rate(CFAR) detection analysis method based on multilook polarimetric whitening filter(MPWF) is proposed in polarimetric synthetic aperture radar(POLSAR) imagery when the clutter obeys Beta distributed texture hypothesis.Firstly,the texture variable in the product model obeys Beta distribution is assumed,and the probability density function(PDF) of MPWF output is derived.Then,the analytical formula of probability of false alarm(PFA) is obtained by integrating probability density function(PDF),and the corresponding CFAR process is designed.Finally,a log-cumulants estimation method based on MPWF is proposed to estimate the texture parameters u and v of Beta distribution.The effectiveness of the new method is verified by simulation data and measured data.Simulation results show that Beta distribution has better goodness of fit on some regions in POLSAR images,and the new method has better CFAR performance compared with the existing methods.The embedded and measured data also show that the new method has better detection performances than the existing methods.