Camera calibration method based on Levy flight and mutation mechanism grey wolf optimization

Abstract

The invention relates to a camera calibration method based on Levy flight and mutation mechanism grey wolf optimization, and the method comprises the following steps: S1, building a nonlinear camera model, and determining calibration parameters; S2, setting the number of grey wolf populations and the maximum number of iterations; S3, acquiring a calibration parameter upper limit and a calibrationparameter lower limit of the camera, and generating a grey wolf position; S4, establishing a target function, and obtaining a back projection error and positions of an alpha wolf, a beta wolf and a gamma wolf; S5, generating a grey wolf intermediate by using a grey wolf optimization algorithm based on Levy flight and a mutation mechanism; S6, updating the position of the grey wolf, and judging whether iteration is continued or not; and S7, obtaining a back projection error according to the target function, wherein the grey wolf position with the minimum back projection error is an optimal calibration parameter. Compared with the prior art, the algorithm can be combined with actual engineering cases, can be accurately and effectively used for multi-dimensional nonlinear problem optimizationsolution, effectively improves the calibration precision, and has good stability and accuracy.

Description

technical field [0001] The invention relates to the field of camera calibration methods, in particular to a camera calibration method based on gray wolf optimization of Levy flight and mutation mechanism. Background technique [0002] Computer calibration is a basic and important step in computer vision, and the quality of the calibration results will also affect the accuracy of subsequent calculations. Therefore, it is very necessary to improve the calibration accuracy. At present, there are two calibration methods widely used and accepted by us: traditional calibration method and self-calibration method. In 2000, a landmark method was proposed by Zhang Zhengyou, based on a two-dimensional checkerboard for calibration. This method improves the calibration object on the basis of the two-step calibration method, and uses the maximum likelihood method to solve each parameter. This method is widely used due to the advantages of low cost, easy operation, and high calibration ...

AI Technical Summary

Problems solved by technology

However, this method requires an accurate calibration object in actual use, and has disadvantages such as sensitivity to initial values, poor convergence, and easy to fall into local optimal solutions.

On the basis of the RAC two-step calibration method, Zhang Zhengyou’s planar calibration method improves the calibration object. It only needs one printed checkerboard to solve the calibration result, which overcomes the shortcomings of the RAC two-step calibration algorithm that requires high-precision calibration objects, but The accuracy of the traditional calibration algorithm has not been greatly improved, and there are still some shortcomings that are the same as the RAC algorithm.

Images