Optimal weighting method of rd equation for multi-base height-ratio spaceborne SAR three-dimensional positioning

Abstract

The invention discloses a multi-base height ratio spaceborne SAR three-dimensional positioning RD equation optimization weighting method, which is based on the RD model to construct the RD equation group under the multi-base height ratio condition; The equations are converted into a matrix form; according to the matrix coefficients, as well as the RD model and the earth ellipsoid equation, calculate the normalized part and the empirical weighted part of the linear equation to determine the weighting matrix; calculate the increment of each iteration according to the weighting matrix, Update the geospatial coordinates of the target point; repeat the iterative update until the iteration meets the judgment stop condition, and the final 3D positioning result is obtained. The invention can not only preserve the direct correlation between the model and the error source, facilitate subsequent error analysis, but also overcome the problems of inaccurate and unstable solutions under the condition of multiple base height ratios.

Description

technical field [0001] The invention relates to a space-borne synthetic aperture radar image three-dimensional positioning technology, in particular to a multi-base height-ratio space-borne SAR three-dimensional positioning RD equation optimization weighting method. Background technique [0002] The three-dimensional positioning and geocoding of each pixel in spaceborne synthetic aperture radar (SAR) images is an important basis for the application of SAR images. To obtain the three-dimensional position of a single SAR image, the support of the earth equation and the DEM library is required, and the multi-angle SAR image of the same scene can be obtained by using two or more angles of observation, and the image can be obtained by solving it in the form of simultaneous positioning equations The three-dimensional position of the target point, the principle is as follows figure 1 shown. Compared with the two-scene SAR three-dimensional positioning, the multi-base height forme...

AI Technical Summary

Problems solved by technology

However, to solve the parameters of the RPC model, it is first necessary to use a rigorous geometric model to establish the geospatial coordinates of several virtual target points and the corresponding image square coordinates, and then fit the model parameters of the RPC model according to the coordinates of these target points, that is, The RPC model is essentially a fitting of a rigorous geometric model, but at this time the specific physical meaning of each parameter in the model is not clear. Therefore, this method loses the direct relationship between the model and the error source, and it is difficult to directly pass the RPC model. Perform error analysis

If the RPC model in the above method is directly replaced by the RD model for solution, since the distance equation of the RD model describes the distance between the radar antenna phase center and the target point at the imaging moment, and the Doppler equation describes the radar antenna phase center The Doppler frequency shift phenomenon produced by the relative movement between the target point and the target point has completely different physical meanings. At the same time, the size of the parameters and their errors in the RD model are also very different. Therefore, the two equations in The numerical difference is very far, which is not conducive to the solution of the equation, especially under the condition of multi-base height ratio, the solution result is very unstable and the error is large

In summary, the existing methods are difficult to take into account the physical meaning of the model and the accuracy of the solution under the condition of multiple base height ratios

Images