基于Marr小波改进的SIFT算法的遥感影像配准  

SIFT Remote Sensing Image Registration Algorithm based on Marr Wavelet

作　　者：张海涛[1] 金燕 刘万军[1] Zhang Haitao;Jin Yan;Liu Wanjun(School of Software,Liaoning Technical University,Huludao 125105,China)

机构地区：[1]辽宁工程技术大学软件学院

出　　处：《遥感技术与应用》2019年第3期622-629,共8页Remote Sensing Technology and Application

基　　金：中国人民解放军总装备部装备预研基金项目(61421070101162107002);辽宁省自然科学基金面上项目(20170540426)

摘　　要：针对遥感图像配准方法中错误匹配点对过多、配准效率低和其他性能,提出了一种基于小波的遥感图像配准方法。首先,利用尺度空间理论下的Marr小波对参考图像和待配准图像进行特征提取,然后利用欧氏距离对参考图像和待配准图像的特征点进行初配准,再根据随机采样一致法,对初配准结果进行精配准。为了验证方法的有效性,选择无人机实时航拍图像、不同时相变化遥感图像以及遥感不同高度的遥感图像。实验结果表明:该方法与SIFT(Scale Invariant Feature Transform)算法以及其他改进SIFT算法相比可以有效剔除错误匹配点对,提高了配准精度,同时提高配准效率两倍以上。该方法可以应用于不同遥感数据源,能够有效地提高配准精度,降低配准时间。For the traditional remote sensing image registration method,there are too many pairs of registration error matching,and the efficiency of registration is low.In order to further improve the accuracy and efficiency of remote sensing image registration,a remote sensing image registration method based on wavelet was proposed.Firstly,the feature extraction of the reference image and the image to be registered using the Marr wavelet in scale space theory.Then use Euclidean distance to perform initial registration of the feature points of the reference image and the image to be registered.Again consistent with the random sampling method,the registration results for early registration for fine.Experimental results show that this method can effectively eliminate false matching points compared with SIFT and other improved SIFT algorithms.Improve registration accuracy,while improving the efficiency of more than double registration.Conclusion:for traditional remote sensing image registration methods,registration mismatches have many pairs of points and the efficiency is low.This paper presents an accurate remote sensing image registration method.The experimental results show that this method can effectively improve the accuracy of registration and reduce the time of registration.

关 键 词：遥感图像配准 Marr小波 欧氏距离 随机采样一致法 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]