结合纹理结构相似度和轮廓相似度的SAR干扰效果评估  

SAR jamming effect evaluation method combining texture structure similarity and image contour similarity

作　　者：刘毅 李姣姣 杨丽丽[1] 徐超[1] LIU Yi;LI Jiaojiao;YANG Lili;XU Chao(China Research Institute of Radiowave Propagation,Qingdao 266600,China)

机构地区：[1]中国电波传播研究所,山东青岛266600

出　　处：《现代电子技术》2023年第17期34-38,共5页Modern Electronics Technique

基　　金：中国电波传播研究所稳定支持科研经费资助项目(A132003W02)。

摘　　要：对SAR干扰效果的评估是衡量抗干扰方法的有效性及SAR系统抗干扰性能的一项重要的综合性指标,文中提出一种结合图像纹理结构相似度和图像轮廓匹配度的SAR干扰效果评估方法(GSIM)。该方法利用图像轮廓匹配度和图像结构相似度构建图像干扰综合评估指标体系,根据指标在不同条件下的重要性进行不同权重的赋值,得到综合的评价结果。为了验证该方法的有效性及准确性,对两组实验图像进行了不同程度的噪声干扰和条纹干扰,通过GSSIM、LKS、GSIM三种评估方法对干扰后的两组实验图像的相似度数据进行对比分析,发现GSIM方法根据不同的干扰环境进行方法组合及权重分配能够对干扰效果进行客观评价,且评价结果更符合人眼的主观感受。该方法为SAR目标的伪装和防护提供了有效的理论支持和技术支撑。The evaluation of SAR(synthetic aperture radar)jamming effect is an important comprehensive index to measure the effectiveness of anti⁃jamming methods and the anti⁃jamming performance of SAR system.In this paper,a method of SAR jamming effect evaluation combining texture structure similarity and image contour matching degree is proposed(GSIM).In this method,the image contour matching degree and the structure similarity are used to construct the image jamming comprehensive evaluation index system,and weights are assigned according to the importance of the indexes to obtain the comprehensive evaluation results.In order to verify the effectiveness and accuracy of this method,two groups of experimental images are interfered by noise and stripes in different degrees.By comparing the similarity data of the two groups of experimental images after jamming with three evaluation methods,namely GSSIM,LKS and GSIM,it is found that the GSIM method can evaluate the jamming effect objectively by combining methods and assigning weights according to different jamming environments,and the evaluation results are more consistent with the subjective feelings of human eyes.This method can provide effective theoretical and technical support for the camouflage and protection of SAR targets.

关 键 词：合成孔径雷达 轮廓相似度 结构相似度 人类视觉系统 干扰效果 抗干扰性能 

分 类 号：TN911.73-34[电子电信—通信与信息系统]