摘要
针对光学和SAR影像存在非线性辐射差异和几何差异造成的匹配精度低问题,提出改进相位特征光学与SAR影像匹配方法。首先,基于相位一致性信息计算最大矩、最小矩,并将两种矩信息进行叠加,接着通过分块的Harris算子在叠加信息上提取稳定的边缘点和角点作为候选特征点;然后,构建多尺度加权累积结构图和方向,提取特征点主方向,构造特征点描述子;最后,利用加权累积结构特征描述子进行初始匹配,通过快速抽样一致性算法计算影像仿射变换矩阵,并利用仿射变换矩阵进行引导的精匹配,降低误匹配率。在光学和SAR影像上对比实验表明,所提出方法在正确匹配点数量、匹配正确率以及耗时上明显占优,表明其有效性。
This paper proposes an improved phase feature optical and SAR image matching method to address the low matching accuracy caused by nonlinear radiative and geometric differences in optical and SAR images.Firstly,based on phase consistency information,the maximum and minimum moments are calculated,and the two types of moment information are superimposed.Secondly,stable edge and corner points are extracted as candidate feature points on the superimposed information using a segmented Harris operator.Then,construct a multi-scale weighted cumulative structure map and direction,extract the main direction of feature points,and construct feature point descriptors.Finally,the weighted cumulative structural feature descriptors are used for initial matching,and the affine transformation matrix of the image is calculated using a fast sampling consistency algorithm.The affine transformation matrix is then used for guided precision matching to reduce the error rate.Comparative experiments on optical and SAR images show that the proposed method has a significant advantage in terms of the number of matching correct points,matching accuracy,and time consumption,indicating its effectiveness.
作者
李建磊
王欢欢
张煦
王纪才
LI Jianlei;WANG Huanhuan;ZHANG Xu;WANG Jicai(School of Artificial Intelligence,Zhoukou Normal University,Zhoukou,Henan 466001,China;School of Economics and Management,Zhoukou Normal University,Zhoukou,Henan 466001,China;Institute of Natural Resources Monitoring and Comprehensive Land Improvement of Henan Province,Zhengzhou 450016,China)
出处
《遥感信息》
北大核心
2025年第2期79-86,共8页
Remote Sensing Information
基金
周口师范学院高层次人才科研启动经费资助项目(72240054)
河南省高等学校重点科研项目(25A520052)
周口师范学院课程思政教育教学改革研究项目(SZJG-2023016)。
关键词
影像匹配
辐射差异
加权累积结构特征
特征匹配
image matching
radiation difference
weighted cumulative structural feature
feature matching
