无RSM参数的反解RFM求解方法  

A method of inverse rational function model solution without rigorous sensor model

作　　者：戴泽源 张立华[1,2] 陈秋 刘翔[1,2] DAI Zeyuan;ZHANG Lihua;CHEN Qiu;LIU Xiang(Department of Military Oceanography and Hydrography,Dalian Naval Academy,Dalian 116018,China;Key Laboratory of Hydrographic Surveying and Mapping of PLA,Dalian Naval Academy,Dalian 116018,China;91001 Troops,Beijing 100841,China)

机构地区：[1]海军大连舰艇学院军事海洋与测绘系,辽宁大连116018 [2]海军大连舰艇学院海洋测绘工程军队重点实验室,辽宁大连116018 [3]91001部队,北京100841

出　　处：《海洋测绘》2023年第2期55-59,共5页Hydrographic Surveying and Charting

基　　金：国家自然科学基金(41871369,41871295,41901320,42071439)。

摘　　要：为实现在无严格传感器模型(RSM)参数的条件下求解反解RFM,提出一种基于正解RFM的反解RFM求解方法。首先通过正解RFM生成虚拟控制格网,建立反解RFM求解流程;然后考虑误差法矩阵病态对求解结果的影响,采用岭参数谱修正迭代与LDLT分解,提升求解的效率和精度;最后以影像立体定位为例,验证了所求解的反解RFM在影像处理中的有效性。实验结果表明:所提方法是对RSM的另一种高精度拟合,在影像立体定位应用中,同正解RFM精度基本相当,且具有更高的立体定位效率,为反解RFM的后续应用提供了良好基础。To solve the inverse RFM without a Rigorous Sensor Model(RSM),a method of inverse RFM solution based on the direct RFM is proposed.Firstly,the virtual control grid is generated by the direct RFM,and the inverse RFM solution flow is established.Secondly,considering the influence of the ill-conditioned normal matrix on the solution results,the ridge parameter and the LDLT decomposition are used to improve the accuracy and efficiency of the solution.Finally,taking image positioning as an example,the effectiveness of the solved inverse RFM in image processing is verified.The experimental results show that the proposed method is a high-precision transfer of the RSM,which is basically the same as the direct RFM in the image positioning,and with higher efficiency.This method provides a solid foundation for the application of the inverse RFM.

关 键 词：有理函数模型 严格传感器模型 反解模型 影像定位 谱修正迭代 LDLT分解 

分 类 号：P23[天文地球—摄影测量与遥感]