Method of correcting drift angle based on deviation statistics of number of overlap pixels between adjacent slices

Abstract

The present invention relates to a method for correcting drift angle based on deviation statistics of the number of overlap pixels between adjacent slices, and the method is as follows: step 1, using an image matching method to obtain the number of the overlap pixels between any adjacent actual image slices and displacement between the slices; step 2 , using assisted data downloaded from a satellite to establish a strict physical imaging model, and calculating to obtain the number of the overlap pixels between any adjacent model image slices and displacement between the slices; step 3, according to the deviation between the number of the overlap pixels between the adjacent actual image slices and the number of the overlap pixels between the adjacent model image slices, calculating to obtain final drift angle error in the assisted data downloaded from the satellite; and step 4, correcting the drift angle downloaded from the satellite. The drift angle error can be calculated by comparing the number, calculated according to the model established by use of the assisted data, of the overlap pixels between the adjacent slices with the number of overlap pixels, obtained by actual image matching, of the overlap pixels between the adjacent slices, and the drift angle in the assisted data downloaded from the satellite can be corrected.

Description

technical field [0001] The invention belongs to the field of aerospace photogrammetry, and realizes a method for correcting the satellite drift angle by counting the deviation of the number of overlapping pixels between adjacent slices when TDICCDs are arranged in a staggered manner or in a zigzag shape inside the camera. Background technique [0002] At present, TDICCD devices are commonly used in high-resolution optical satellites, and their imaging quality is good. However, due to the small field of view and large casing, multiple TDICCDs are often arranged staggered on the focal plane to expand the field of view. Insufficient control precision results in missing scans, and a certain number of pixels overlap between adjacent slices. In the auxiliary data downloaded by the satellite, there are three attitude angles, which are: pitch angle, roll angle and yaw angle. In TDICCD, the yaw angle is the drift angle. The purpose of deflecting the satellite platform by an angle (t...

AI Technical Summary

Problems solved by technology

But in fact, due to the measurement error of the satellite attitude angle (that is, the attitude determination error), especially the measurement error of the bias angle is relatively large, this leads to the fact that the actual number of overlapping pixels between slices is different from the overlapping image calculated by modeling with auxiliary data. Inconsistent arity (there is a bias)

After calculation, in the case of a distance of 2600 pixels between adjacent slices, a deviation angle measurement error of 0.06 degrees will result in a deviation of the number of overlapping pixels between slices of 2600*tan(0.06)=2.7 pixels

[0004] In the case of TDICCD push-broom imaging, the roll angle and pitch angle errors in the satellite attitude angle lead to geometric positioning errors in the vertical and along-track directions of the image respectively (this problem can be solved by simple coordinate translation), while the drift angle error Reduce the internal geometric accuracy of the stitched image (this problem is difficult to solve)

Images