Full-model lens distortion correction method adopting dual-characteristic joint measurement

Abstract

The invention provides a full-model lens distortion correction method adopting dual-characteristic joint measurement, which comprises the following steps of: carrying out sub-pixel-level point coordinate detection on a detected image, and determining a distortion center by utilizing a least square method according to a relation equation of a basic matrix and the distortion center; calculating an inclination angle and a transformation factor of a pixel unit on the detected image to form an inverse mapping matrix in a digital image discrete imaging process, and converting a non-standard pixel unit into a square pixel unit; calculating the coordinates of the blanking points according to the significant features of the perspective projection and the homography matrix, and updating the fitted distortion straight line by using the blanking points; and on the basis of the full distortion model, establishing a function constraint relationship with a distorted straight line according to geometric constraints of the blanking points, and minimizing a straight line residual error by using an optimization method to obtain a distortion parameter, an inclination angle and a transformation factor, thereby realizing accurate distortion correction of the camera. According to the method, lens distortion correction can be rapidly and accurately realized, the correction effect is good, and the robustness is high.

Description

technical field [0001] The invention relates to the technical field of close-range measurement and digital image processing, in particular to the technical field of lens distortion correction, specifically a full-model lens distortion correction method using dual-characteristic joint measurement, and using checkerboard patterns to realize accurate lens distortion correction. Background technique [0002] In recent years, with the rapid development of the field of artificial intelligence, more and more researchers have devoted themselves to the research in this field. Since machine vision plays an important role in artificial intelligence, zoom lenses are widely used in aerial photography, tracking and surveying. Lens distortion is prevalent in optical systems, especially in fisheye and wide-angle lenses, due to lens imperfections and misalignment of optical components. When the lens has non-negligible distortion, using an ideal pinhole model may lead to high measurement err...

AI Technical Summary

Problems solved by technology

[0006] At present, there are many deficiencies in the traditional lens distortion correction method, for example: (1) The image center or principal point coordinates are defaulted as the distortion center, and the distortion center is not calculated accurately, resulting in the consequences that it is impossible to use distortion Parameter compensation to correct, and it is impossible to use the external orientation parameters to correct; (2) The parallelogram pixels existing in the projective transformation are not considered, let alone processed, and the linear imaging process in the projective geometry cannot be guaranteed; (3 ) The algorithm based on the straight line theory only focuses on radial distortion or eccentric distortion, and does not perform lens distortion correction under the complete distortion model

Images