基于可变参数化几何模型的SAR图像飞机目标特征提取方法  

Feature Extraction of Aircraft Targets in SAR Image Based on Parametric Geometric Model

作　　者：陈玉洁[1] 赵凌君[1] 匡纲要[1] 

机构地区：[1]国防科技大学电子科学与工程学院,长沙410073

出　　处：《现代雷达》2016年第10期47-53,共7页Modern Radar

摘　　要：基于飞机目标的轮廓特点和成像特性,提出了一种基于可变参数化几何模型的合成孔径雷达(SAR)图像飞机目标特征提取方法。首先,利用飞机目标的先验知识构造一个描述飞机目标外形轮廓的参数化模型;然后,对于输入的实测飞机目标切片图像,构建目标函数来度量模型与实测图像中目标区域的拟合程度,通过遗传算法求解最优参数;最后,在最优参数模型的基础上计算目标的几何特征。基于仿真和实测数据的实验结果验证了该方法的有效性。Based on the contour features and imaging characteristics of aircraft targets, a method of feature extraction for aircraft tar-get in SAR images based on the variable parametric geometry model is presented. Firstly, a parametric model of aircraft target con-tour is constructed based on the prior knowledge of aircraft target in this method. Then, the objective function is constructed to measure the fit degree of the target area in the SAR image and model, and solve the optimal parameters by genetic algorithm. Fi-nally, the geometric feature of target is calculated on the basis of the optimal parameter model. Experimental results based on simu-lated and measured data demonstrate the effectiveness of the proposed method.

关 键 词：合成孔径雷达 飞机目标 特征提取 参数化几何模型 

分 类 号：TN957.52[电子电信—信号与信息处理]