三线阵影像外方位元素平滑方程自适应光束法平差

Self-adaption Bundle Adjustment of Three-line Array Image with Smoothing Equation of Exterior Orientation Elements

外方位元素数据是影响光学卫星影像几何定位精度的重要因素,其精度不仅取决于姿态和轨道测定系统的精度,还与卫星平台稳定度紧密相关。在实际卫星工程中,姿态测定系统存在突变现象,导致处理航线中上下视差发生较大变化。本文在全三线交会EFP光束法平差理论基础上,提出在光束法平差中采用外方位元素自适应分段平滑策略及其数学模型。即根据立体影像上下视差差分数据,对立体影像进行自动分段;在此基础上,对不同段的外方位元素平滑方程赋予不同权值,共同参与光束法平差。通过实际卫星影像试验表明,该方法可有效抑制卫星姿态测定系统突变对立体影像无控定位精度的影响,尤其是高程精度改善显著。

Exterior orientation elements is a key factor to effect the location accuracy of optical satellite images,and its accuracy is depended on not only the attitude and orbit determination system,but also the stability of satellite platform.However,there is a mutation phenomenon in satellite engineering,which will lead to a big change of vertical parallax in processing strip.In this paper,based on the theory of EFP bundle adjustment of all three-line intersection,the strategy and mathematical model of self-adaption bundle adjustment with smoothing equation of exterior orientation elements are proposed.Firstly,the image is segmented automatically according to the vertical parallax difference data of the stereoscopic image.Then,the different weights of smoothing equation of external orientation element are given with different image segments,and participated in bundle adjustment.Finally,the experiments are performed and the results show that the proposed method can effectively restrain the influence of the satellite attitude observation mutation to the accuracy of positioning without control of stereo images,especially the elevation accuracy improved significantly.