A fisheye image correction method based on multi-step correction

Abstract

The present invention introduces a fisheye image correction method based on multi-step correction. In view of the deficiencies such as image amplification and edge information loss caused by the existing conventional correction model, a correction model is constructed using the idea of ​​multi-step correction. The correction model constructed The key point is: establish the fisheye image size scaling factor based on the polynomial model, use the fisheye image size scaling factor to perform the first correction of the fisheye image; use the neighborhood interpolation method to interpolate the first correction image; establish Based on the division model of the backward mapping method, the division model is used to perform a second correction on the first corrected image after interpolation processing to obtain a second corrected image. The correction model of the fisheye image solves the problems of missing edge information of the corrected image, image expansion, incomplete distortion correction, etc., so that most of the information of the fisheye lens is completely preserved, and the blurred image edge is improved. .

Description

technical field [0001] The invention relates to the fields of image processing and machine vision, in particular to a fisheye image correction method based on multi-step correction. Background technique [0002] Fisheye lens is a special wide-angle lens, its field of view is close to, equal to or even greater than 180 degrees, and can image objects in the hemispherical airspace or even the super-hemispherical airspace within a limited range of the image plane. From the perspective of optical design, the lens used in the fisheye lens has a large spherical curvature and is closer to the imaging plane. This special structural feature and imaging feature, on the one hand, enables the fisheye camera to obtain a large field of view, and can be widely used in real-life scenes that require a large field of view, such as robot navigation, video conferencing, real-time monitoring, panoramic photography, and astronomical observation. On the other hand, due to the introduction of large...

AI Technical Summary

Problems solved by technology

From the current engineering practice, although these algorithms can achieve distortion correction to different degrees, the correction of large distortion images such as fisheye images is often local, that is to say, there is loss of image edge information, and image local Disadvantages such as magnification, which violates the original intention of people using fisheye lenses, weakens the advantages of wide-angle lenses to a certain extent

Images