Method and system for correcting fitting error of general imaging model of satellite image

Abstract

The invention provides a method and system for correcting a fitting error of a general imaging model of a satellite image. The method comprises the following steps: establishing a three-dimensional space grid based on a strict geometric imaging model; carrying out the re-imaging correction grid establishment and image point grid coordinate correction, including the step of fitting attitude metadata by using a cubic polynomial, establishing an attitude polynomial corresponding to an ideal image, setting a re-imaging correction grid, calculating an initial value by using attitude transformationre-imaging, and calculating a correction value corresponding to an image point grid by using re-imaging correction grid interpolation; fitting residual statistical test and re-imaging correction gridoptimization are carried out, including the steps of utilizing an object space three-dimensional space grid and a corrected image point grid to calculate RPC parameters through a least square method,calculating image point coordinates corresponding to the re-imaging correction grid through the RPC parameters, calculating the fitting residual error of the universal imaging model through the re-imaging correction grid and the image point coordinates, and calculating the correction amount of the re-imaging correction grid through the fitting residual error corresponding to the grid points, thereby achieving the optimization of the re-imaging correction grid.

Description

technical field [0001] The invention belongs to the technical field of remote sensing, and in particular relates to the compensation and correction of the fitting error of a satellite image rational polynomial general imaging model. Background technique [0002] The satellite image imaging model expresses the mathematical relationship between the coordinates of the image point and the object space coordinates of the corresponding target point, which is the basis of the geometric processing of the satellite image. Geometric imaging models of satellite images can be divided into two types: Rigorous Geometric Sensor Model (RSM) and Rational Polynomial Function Sensor Model (RFM). Due to the advantages of simple form, strong versatility, and independent of the sensor structure, the general imaging model based on rational functions has become the most widely used imaging model, and is widely used by satellite image providers, and almost all satellite images provide rational funct...

AI Technical Summary

Problems solved by technology

[0007] (3) The composition of the three-dimensional model is complex;

[0008] (4) Image coordinate conversion is troublesome, especially when ground coordinates are projected to image coordinates, it is necessary to determine which sub-block the ground point is located in through iterative calculations, and use the corresponding RPC parameters

[0009] In order to achieve generalized and convenient photogrammetry processing, it is necessary to carry out overall RPC modeling on the long strip image to generate the RPC of the entire segmented image, which leads to a relatively large fitting of the relatively strict geometric imaging model of RFM Error, which affects the positioning accuracy and application of general imaging models

Images