Block adjusting method of satellite image of three-line array

Abstract

The invention relates to a block adjusting method of a satellite image of a three-line array. The method comprises the following steps: (1) reading the satellite image and storing coordinate information of all types of points as corresponding point documents; (2) reading the documents to obtain an RPC (Remote Position Control) model as well as ground coordinates and image coordinates of a control point and a check point and image point coordinates of a connecting point; (3) determining the initial value of a ground point coordinate and image space conversion model; (4) constructing an error equation point by point and normalizing; (5) solving a corrected parameter and coordinates of a ground point for a normal equation by means of spectral correction by iterative estimation; (6) calculating the object space precision achieved after adjusting iteration at the time; (7) calculating the image space precision after adjusting iteration at the time, and determining the weight P of a coordinate observation value of an iterative image point at the next time by means of an information diffusion model through image space residue; (8) when translation parameters in orientation parameters are less than a threshold value, finishing adjusting iteration; if the predetermined condition is not satisfied, continuing iterating till iteration is finished; and (9) when the whole balancing iteration is finished, outputting a precision report on the most adjusting object space and image space precisions.

Description

technical field [0001] The invention relates to the technical field of photogrammetry, in particular to a three-line array satellite image block adjustment method. Background technique [0002] High-resolution push-broom sensors can acquire high-resolution remote sensing images. Current high-resolution satellite remote sensing images are basically acquired based on this special imaging method. Since the beginning of the 21st century, more and more high-resolution commercial satellites have been successfully launched, providing users with a variety of data source options. However, due to the different design of each satellite sensor, the practicability of the acquired satellite images is greatly limited. Therefore, in order to establish a strict one-to-one correspondence between image coordinates and ground coordinates, it is necessary to establish an imaging geometric model of the image. At present, the commonly used imaging geometric models are mainly divided into two typ...

AI Technical Summary

Problems solved by technology

The RPC-based model has the advantages of high simulation accuracy, good versatility, convenient application, and small amount of calculation. Corresponds to rigorous models with physical meaning

Images