基于单通道SAR图像序列特征值分解的动目标检测方法  

The Method of Moving Target Detection Based on the Eigen-Decomposition of the Single-Channel SAR Image Sequence

作　　者：刘书君[1] 袁运能[2] 毛士艺[2] 

机构地区：[1]重庆大学通信工程学院,重庆400044 [2]北京航空航天大学电子信息工程学院,北京100191

出　　处：《电子与信息学报》2009年第8期1876-1880,共5页Journal of Electronics & Information Technology

摘　　要：该文提出了一种基于单通道图像序列间协方差矩阵分解的动目标检测方法。首先给出基于方位频谱划分获取子图像的处理过程,分析了子孔径划分在图像序列间所产生的误差来源,结合二维自适应方法对幅度和相位上存在的误差同时校正,实现了子图像间的配准,构造出类似于多通道的子图像。在此基础上,结合多通道杂波抑制的思想,详细分析了两子孔径间协方差矩阵特征值分解实现目标与杂波分离的原理,并针对在图像域估计采样协方差引起的精度与目标能量损失之间的矛盾,提出了在距离多普勒域的改进处理。最后,经过仿真实验验证了该方法的有效性。The method of moving target detection based on eigen-decomposition of the covariance matrix is proposed for single-antenna airborne SAR. First the paper points out that the sub-image is obtained through frequency processing. The imaging difference of sub-images as the result of the sub-aperture division is rectified using the adaptive 2D-calibration. The magnitude and phase difference can be corrected at the same time and the sub-images like the multi-channel can be obtained. Then the developed method combines the concept of multi-channel samples covariance matrix. The fundament of the separation between clutter and moving target based on the decomposition of the sub-image sample covariance matrix is analyzed. The estimation of the sample covariance matrix in image brings the inconsistency between the accuracy and the decrease of the target＇s energy. The improved method of estimation is calculated in range-doppler domain is proposed and the moving target is indicated based on the second eigenvalues. Simulation results demonstrate that the method is effective.

关 键 词：合成孔径雷达 运动目标检测 协方差矩阵特征值分解 

分 类 号：TN957.52[电子电信—信号与信息处理]