摘要
在获取遥感影像数据过程中,因受到传感器姿态、地面起伏、地球曲率等影响,遥感影像会出现几何失真,需要对遥感影像进行几何纠正.多项式几何纠正模型在利用最小二乘估计法求解多项式系数时,会遇到病态方程问题.利用主成分分析模型求解多项式系数,结合实例遥感影像数据精准求解多项式系数,实现遥感影像的几何纠正.试验结果表明,主成分分析的几何纠正误差比最小二乘估计法的误差小,解决了病态方程问题,为遥感影像数据的几何纠正提供了技术解决方案.
In the process of acquiring remote sensing image data,geometric distortion of remote sensing images occurs duet to influences of sensor attitude,ground undulation,and earth curvature.Therefore,geometric correction of remote sensing images is required.The polynomial geometric correction model is more likely to encounter ill-conditioned equation problems when least squares estimation is used to solve polynomial coefficients.To solve this problem,this paper uses principal component estimation to solve the polynomial coefficients,combined with example data,and compares the geometric correction results of least squares estimation and principal component estimation.The experimental results show that the geometric correction error of the principal component estimation is smaller than that of the least squares estimation,which solves the problem of ill-conditioned equations and provides a technical solution to geometric correction of remote sensing image data.
作者
王海青
WANG Hai-qing(Suzhou Institute of Construction&Communications,Suzhou 215104,China)
出处
《南京工程学院学报（自然科学版）》
2020年第4期84-88,共5页
Journal of Nanjing Institute of Technology(Natural Science Edition)
关键词
多项式纠正模型
主成分分析
病态方程
几何纠正
最小二乘估计法
polynomial correction model
principal component analysis
ill-conditioned equation
geometric correction
least square estimation
作者简介
王海青,硕士,讲师,研究方向为大地测量学与测量工程.E-mail:269487535@qq.com。
