Camera self-calibration method

Abstract

The invention provides a camera self-calibration method, which comprises the following steps: step A1, aiming at multiple images shot at multiple angles at the same scene, extracting two-dimensional quasi-dense points through affine-change-based regional expansion algorithm, and obtaining two-dimensional matching points of the two-dimensional quasi-dense points on neighborhood images by tracking; step A2, utilizing two-dimensional quasi-dense points and the two-dimensional matching points of the two-dimensional quasi-dense points on the neighborhood images, obtaining the three-dimensional positions of the quasi-dense points through SFM algorithm, and obtaining camera parameters and a three-dimensional scene structure by restoring; step A3, carrying out iterative optimization of the three-dimensional scene structure and the camera parameters; step A4, combining the camera parameters to select a neighborhood image set corresponding to each image; step A5, resampling the quasi-dense points and the corresponding matching points; and returning to the step A2, and executing the step A2 to step A5 multiple times in a looping way, wherein the last loop ends until the step A3. Through the invention, the accuracy and robustness of calibration results are improved.

Description

technical field [0001] The invention relates to the technical fields of digital image processing and computer vision, in particular to a camera self-calibration method. Background technique [0002] Camera calibration is an important research content in the field of computer vision. It is an essential step to extract 3D spatial information from 2D images. It is widely used in 3D reconstruction, navigation, visual monitoring and other fields. In the pinhole camera model, the process of solving the projection matrix P is called camera calibration; correspondingly, the process of solving the intrinsic parameter (Intrinsic parameter) matrix K is called internal calibration, and the process of solving the extrinsic parameter (Extrinsic parameter) ) The process of R and t is called external calibration. [0003] There are three types of camera calibration in a broad sense: traditional calibration methods, active vision-based calibration methods, and self-calibration methods. The...

AI Technical Summary

Problems solved by technology

However, in this kind of expansion algorithm, a greedy matching strategy is adopted, and there is no stricter filtering mechanism, so there must be a large number of mismatches, which will affect the accuracy of quasi-dense points

In addition, the seed point expansion algorithm used in it is not suitable for wide baseline situations, so there are still deficiencies in practical applications

[0007] Second, the Bundle adjustment algorithm is an optimization algorithm widely used in the calibration process and 3D modeling field. It optimizes the camera parameters and the structure of the scene by minimizing the reprojection error as the objective function, but when the input image When the number of and 3D points are large, its overhead becomes very large, and even the optimization fails

Images