| Geometric positioning of linear satellite images,as one of the keys in producing geoinformation products,plays an important role in development of national economic.For the situation of lack of actual measured control information,a method of geometric positioning with multi-source data is proposed,.Based on control data and positioning model,the control points extracted from existing geographic data(named as auxiliary control points)are used for improving accuracy in geometric positioning without adding actual measured control points,meanwhile,total least-squares(TLS)is introduced for rational function model(RFM)to make the results of geometric positioning more reasonable.The main work and innovations of this paper are as follows:1.A method of geometric positioning based on RFM with multi-source data is proposed.The auxiliary control points extracted from high-precision images and digital elevation model(DEM)are used in geometric positioning based on RFM with few or no actual control points.The experimental results show that the method is feasible and effective to improve the accuracy of geometric positioning.2.For the reason that high-precision images with large size are difficult to be acquired,and may cause large amount of data and calculation,the method of geometric positioning with small size high-precision images for large size satellite images is studied.The experimental results show that concentratied auxiliary control points are capable to improve accuracy of geometric positiong.3.Considering the random errors in coefficient matrix of image points observation equation in error equation,the Errors-in-variables(EIV)model is introduced in RFM geometric positioning.The TLS esprit which is suitable for the block adjustment of RFM is derived then applied to the geometric positioning with multi-source data to make the positioning more rigorous and more reasonable.The experimental results show that the method has effects on improving the positioning results. |
PDF Full Text Request