SAR图像目标方位角估计与分析

SAR成像对目标方位角特别敏感,方位角估计的准确性会直接影响目标的分类和识别.目标方位角估计方法主要有包络盒准则法、轮廓线变换法和主轴原则法.每种方法都有其局限性,且估计范围是0, 180],可是目标方位角实际范围是0, 360].为了能获得任意角度且相对精确目标方位角,提出了一种综合SAR图像目标方位角估计方法.该方法融合了的单一估计法的优点,并利用目标的相关信息,突破了以往仅从图像的角度估计方位角的缺点.实验表明这种方法是有效的、可行的.

Pose Estimation and Analysis of SAR Image Targets

Synthetic Aperture Radar(SAR) imaging is especially sensitive to pose,and the accurate estimation of pose directly influences the classification and recognition of targets.The main methods of pose estimation include target-enveloped box criterion,target contour line transform and principal axis.Every method all has its limitation and the estimated range is 0,180 ,in fact,the practical pose of a target is 0,360 .In order to obtain random and relatively accurate target pose,a hybrid SAR image pose estimation method was proposed.The method fuses the advantage of every algorithm and fully uses the relative information of target,overcoming the shortcoming of estimating pose only from image viewpoint.Experiments verify that the method is effective and feasible.