基于高斯低通滤波算法的合成孔径雷达图像快速配准模型  

Fast Synthetic Aperture Radar Image Registration Model based on Gaussian Low-Pass Filtering Algorithm

作　　者：江明达 宋相杨 张志雄 赵阿珠 刘忠会 JIANG Mingda;SONG Xiangyang;ZHANG Zhixiong;ZHAO Azhu;LIU Zhonghui(Unit 70 of unit 32286 of the Chinese People’s Liberation Army,Tieling 112000,China)

机构地区：[1]中国人民解放军32286部队70分队,辽宁铁岭112000

出　　处：《电声技术》2022年第4期92-95,共4页Audio Engineering

摘　　要：为了提高合成孔径雷达图像配准的正确匹配数与速度,提出基于高斯低通滤波算法的合成孔径雷达(Synthetic Aperture Radar,SAR)图像快速配准模型。在图像匹配前,利用低通滤波器归一化SAR图像背景,为后续图像快速匹配奠定基础。在此基础上,通过高斯低通滤波算法滤除图像的散斑噪声,最终实现图像的快速配准。仿真实验结果表明,所提出的方法可以有效提高配准正确数量,且配准运行时间低于8 s,运行稳定,可以为图像配准提供一定基础依据。In order to improve the correct matching number and speed of synthetic aperture radar image registration, a fast registration model of Synthetic Aperture Radar(SAR) image based on Gaussian low-pass filtering algorithm is proposed. Before image matching,the background of SAR image is normalized by low-pass filter, which lays the foundation for subsequent image fast matching. On this basis, the speckle noise of the image is filtered by Gaussian low-pass filtering algorithm, and finally the fast image registration is realized. The simulation results show that the proposed method can effectively improve the correct number of registration, and the registration running time is less than 8 s, the operation is stable, and can provide a certain basis for image registration.

关 键 词：图像配准 高斯低通滤波 合成孔径雷达 背景归一化 散斑噪声 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]