基于线特征的rpc立体定位系统误差补偿模型-同济大学学报.doc

基于直线特征的有理函数模型定位精度改善方法 叶勤1，2，张小虎1，3，刘世杰1，谢惠洪4 （1.同济大学 测量与国土信息工程系，上海 200092；2.同济大学 遥感与空间信息技术研究中心，上海 200092；3.同济大学 现代工程测量国家测绘局重点实验室，上海 200092；4.上海市测绘院，上海 200063） 摘要：探讨了利用已知直线特征进行有理函数模型定位误差在物方进行补偿的方法. 推导了两种物方补偿模型（平移加主因子比例和一阶仿射变换模型）下，基于已知直线特征进行补偿模型参数解算的数学模型. 给出了利用已知直线特征进行有理函数模型定位误差补偿的解算方案. 实验表明：基于线特征的补偿方法是可行的，并且平移加主因子比例模型的精度好于基于控制点的方法，也好于一阶仿射变换模型的补偿效果；但对于一阶仿射变换模型，该方法的补偿效果要比基于控制点的方法差。 关键词：成像模型；有理多项式函数；直线特征；平移加主因子比例模型；一阶仿射变换模型 中图分类号：p236 文献标识码：A 文章编号：0253-374X（2010）02- The Refine Method of Positioning Accuracy of Rational Function Model Based on Straight Line features YE Qin1,2, ZHANG Xiaohu1,3, LIU Shijie1, XIE Huihong4 (1.Department of Surveying and Geo-Informatics, Tongji University, Shanghai 200092, China; 2.Research Center for Remote Senesing and Spatial Information, Tongji University, Shanghai 200092, China; 3.Key laboratory of Advanced Engineering Survey of SBSM, Tongji University, Shanghai 200092, China; 4.Shanghai Municipal Institute of Surveying and Mapping, Shanghai 200063, China) Abstract: This paper discussed the method which refines the RFM-based stereo positioning in object space based on known straight line features. For the two refine models (the Shift and main factor model and the Affine model), the formulas were derivated to solve the refine model parameters based on known straight line features. And the solution using known straight line features was given to compensate the RFM positioning error. The experimental results show that it is feasible to refine the RFM-based stereo positioning based on line features; and for the Shift and main factor model, the accuracy of using the straight line features is better than that using points, but for the Affine models, the accuracy of using straight line features is worse than that using points; furthermore, under the same condition, the accuracy of the former is also better than the latter. Key words: Imaging model; Rational function model; Straight line features; Shift and main facto