A fisheye image correction method and device based on coordinate transformation

Abstract

The invention discloses a fisheye image correction method and device based on coordinate transformation. A pixel point P' is arbitrarily selected from the fisheye image as the center point for correction of the fisheye image, and is calculated by the imaging model of the fisheye camera. The spherical coordinates of the incident point P of P′ in the fisheye hemisphere, any plane parallel to the tangent plane that is tangent to the fisheye hemisphere with point P as the tangent point is taken as the plane where the corrected image is located, and the intersection point Q between the plane and the ray OP Establish the virtual imaging surface of the corrected image for the center point of the corrected image, calculate the coordinates of the pixel points corresponding to each drawn point in the fisheye image one by one, and calculate the pixel value of the pixel corresponding to the drawn point in the fisheye image The pixel values ​​of the plotting points in the corrected image are plotted in the corrected image to obtain the corrected image of the fisheye image. The method and device of the present invention perform multi-directional real-time correction on the high-definition fisheye camera, so as to facilitate the observation of details in different areas in the fisheye image.

Description

technical field [0001] The invention belongs to the field of video surveillance, and in particular relates to a fisheye image correction method and device based on coordinate conversion. Background technique [0002] In many application scenarios, users hope to monitor a larger range in the monitoring screen of a video surveillance camera, so super wide-angle lenses have been used more and more. Among them, fisheye lenses have the ability to obtain more than 180 degrees The image of the field of view realizes monitoring without dead angle, and is widely used in video monitoring. Although the fisheye camera can obtain images with a large field of view, the images taken by the fisheye camera have serious distortions and do not conform to the observation habits of the human eye. Therefore, it is necessary to correct the distortion of the images taken by the fisheye camera so that people The monitoring area of ​​the eye camera is observed. The distortion correction technology ...

AI Technical Summary

Problems solved by technology

[0004] Although the above method can correct the fisheye image, the method first needs to map the fisheye image into a spherical image, and the calculation amount of mapping the fisheye image into a spherical image is very large, and when the fisheye image is corrected by this method , does not need to use all the pixels of the fisheye image, so the efficiency of this method is low

Images