Method utilizing curve-fitting to calibrate radial distortion of camera

Abstract

The invention discloses a method utilizing curve-fitting to calibrate radial distortion of a camera, belonging to the technical filed of image processing. The method comprises the following steps: adopting a checker board as a template, collecting the image of the template and finding a distortion center, detecting a linear projective point set and transferring the original point of an image plane to the distortion center, utilizing a round fitting method to obtain the equation of the arc of the original linear projection to obtain radial distortion parameters, and finally utilizing a traditional plane calibration-based method to estimate the inner parameters of other cameras. The invention avoids the complexity of an iterative method and the limitation of falling into a local optimum value, and the curve-fitting method is utilized to obtain the radial distortion parameters and is simple, convenient and effective; and on the other hand, the method of the invention separates the radial distortion parameters from the internal parameters of other cameras, thereby being more flexible than the existing method.

Description

technical field The invention relates to a method in the technical field of image processing, in particular to a method for correcting radial distortion of a camera by curve fitting. Background technique Radial distortion of camera lens widely exists in wide-angle cameras, fisheye cameras and cheap cameras with short focal length, so radial distortion is an important issue to be considered in analyzing digital images. At present, there are many methods for restoring distortion parameters. R. Swaminathan et al. published the paper "Nonmetric calibration of wide-angle lenses and polycameras" (wide-angle lens and Non-metric calibration of multiple cameras) for distortion correction based on the so-called plumb line theory, that is, straight lines in space are still projected as straight lines under the ideal pinhole projection transformation. According to this principle, the projected point set of the known spatial straight line is detected first, and then the objective func...

AI Technical Summary

Problems solved by technology

Although this method has no requirements on the scene,

Images