基于模型简化和像素加权的最小二乘匹配方法  

Least square image matching method based on model simplification and pixel weighting

作　　者：杨阿华 张强 汪世辉[1] 张希 YANG Ahua;ZHANG Qiang;WANG Shihui;ZHANG Xi(Beijing Institute of Tracking and Telecommunications Technology,Beijing 100094,China)

机构地区：[1]北京跟踪与通信技术研究所,北京100094

出　　处：《航天工程大学学报》2025年第1期89-95,共7页

摘　　要：针对传统最小二乘影像匹配方法稳定性较差、不易收敛的问题,提出了一种基于模型简化和像素加权的最小二乘影像匹配方法。将误差模型中的亮度偏差与增益参数进行消除以简化模型,在基于误差模型构造约束方程组时,对各像素用高斯窗进行加权,同时对R、G、B3个颜色通道分别建立约束方程并联立解算。通过对10个航空影像对的稠密同名点进行最小二乘匹配,各像对采用本方法的收敛率和相关系数较传统方法均有所提升,处理时间平均降低14%,证明本方法的稳定性、精度、效率均优于传统方法。In view of the poor stability and difficulty in convergence associated with the traditional least square image matching,a least squares image matching method based on model simplification and pixel weighting is proposed.Firstly,the deviation and gain parameters of brightness in the error model are eliminated to simplify the model.In constructing the constraint equation group based on the error model,each pixel is weighted with a Gaussian window,and constraint equations are established separately for the R,G,and B color channels and solved simultaneously.Least square matching is performed on dense corresponding points from 10 aerial image pairs.The experimental results show that the convergence rate and correlation coefficient of each image pair using this method are improved compared to traditional methods,and the processing time is reduced by 14%in average.This indicates that the stability,accuracy,and efficiency of this method are better than traditional method.

关 键 词：最小二乘匹配 偏差与增益 仿射变换 像素加权 

分 类 号：P236[天文地球—摄影测量与遥感]