Large distortion checkerboard image angle point detection method and device

Abstract

The invention discloses a large distortion checkerboard image angle point detection method. The method comprises steps that a center point of a communication area of a large distortion checkerboard image is taken as an origin to establish a rectangular coordinate system; contour points consistent with preset conditions are selected according to coordinate values of contour points of the communication area, and the selected contour points are defined as angle point candidate points; taking the center point as the origin, the rectangular coordinate system is turned according to a preset turningmode, and new angle point candidate points are re-defined; after the turning frequency of the rectangular coordinate system reaches the preset turning frequency, the frequency of each contour point defined as the angle point candidate point is acquired, and the angle point candidate points in preset quantity are taken as angle points of the communication area. The invention further discloses a large distortion checkerboard image angle point detection device. The method is advantaged in that the method is simple, convenient and rapid, and the angle points of the single communication area of thelarge distortion checkerboard image can be detected.

Description

technical field [0001] The invention relates to the field of digital image processing, in particular to a method and device for detecting corner points of a checkerboard image with large distortion. Background technique [0002] The 360° surround view panoramic assisted driving system is a hot research issue today. Although the fisheye camera used in it obtains a large field of view image, it also introduces serious distortion. These severely distorted images cannot meet the normal visual needs of people. In order to enhance the visual effect of the image, it is necessary to calibrate the fisheye distorted image and detect the corner information, so as to construct its spatial correspondence with the real world coordinate system. Checkerboard images are often used in fisheye camera calibration. The corner detection methods of large-distortion checkerboard images mainly include SIFT (Scale-invariant feature transform, scale-invariant feature transform) and Martin (Flajolet-Ma...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of the present invention is to provide a large distortion checkerboard image corner point detection method, device and computer storage medium, aiming to solve the problem that the existing large distortion checkerboard image corner point detection algorithm has a large amount of calculation and cannot detect the corner point of a single connected area. The problem

Images