摘要
目前基于多幅极化SAR图像融合的遥感应用受到了广泛关注.图像配准是图像融合应用的一个关键预处理步骤.典型的SAR-SIFT配准算法不能利用极化SAR图像的目标极化信息.本文提出Wishart-SIFT配准方法,即利用Wishart梯度代替SAR-SIFT中的比值梯度,并基于此梯度建立改进的SIFT配准算法,从而充分利用极化SAR图像的目标极化信息.我们在RADARSAT-2和AIRSAR的数据集上分别实验验证了Wishart-SIFT方法在不同视角和不同频段极化SAR图像下的配准效果,结果显示Wishart-SIFT方法在配准精度和稳定性方面均优于SAR-SIFT方法.
Applications based on multipolarimetric SAR images have received increasing attention recently.The existing SAR.SIFT algorithm cannot completely utilize the polarimetric scattering information.Thus,in this paper,the Wishart.SIFT registration method is proposed.Wishart gradient is used to calculate the image gradients,and a novel registration method based on a modified SIFT algorithm is proposed to utilize the polarimetric information.Experimental results with the RADARSAT-2 and NASA/JPL AIRSAR data demonstrate that the proposed method outperforms the SAR.SIFT method,especially in the registration between images of different views and frequencies.
作者
王洪淼
杨健
殷君君
Hongmiao WANG;Jian YANG;Junjun YIN(Department of Electronic Engineering,Tsinghua University,Beijing 100084,China;School of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083,China)
出处
《中国科学：信息科学》
CSCD
北大核心
2022年第4期687-699,共13页
Scientia Sinica(Informationis)
基金
国家自然科学基金(批准号:62171023)
高分辨率对地观测重大专项航空观测系统(批准号:30-H30C01-9004-19/21)资助项目。
