Marker positioning method and system of calibration plate image

Abstract

The present invention discloses a marker positioning method and system of a calibration plate image. The method includes the following steps that: image pyramid layering is performed on the calibration plate image; binarization is performed from a topmost layer of image, a calibration plate area and a marker contour are searched in the binary image, if no calibration plate areas and marker contours are found out, the calibration plate area and the marker contour are further searched in a next layer of image until search is performed in the last layer or the calibration plate area and the marker contour are found out; and the bounding rectangle of the marker contour in the calibration plate area is calculated in the pyramid image of the searched calibration plate area, the bounding rectangle is mapped into the calibration plate image, so that a region of interest containing a marker is obtained, and the center coordinates of the marker are calculated. Therefore, contour search is performed from the top-layer low-resolution image, so that the calibration plate area and the marker contour can be found out, and the region of interest containing the marker can be positioned quickly and accurately, and positioning efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a method and a system for locating markers on a calibration plate image. Background technique [0002] In the era of Industry 4.0, precision measurement and detection based on machine vision is a very important research field, whether it is in traditional manufacturing, 3C electronics industry or aerospace field, it has very broad application prospects. Camera calibration is the basis for machine vision to obtain precise measurement data. Camera calibration is the process of solving the internal and external parameters of the camera through the correspondence between the world coordinate system and the image coordinate system. The accuracy of the actual measurement data depends on the internal parameters of the camera. The accuracy of parameters and external parameters, then in the final analysis, a set of accurate correspondence between the world coordinate system and the...

AI Technical Summary

Problems solved by technology

[0004] At present, most of the existing landmark positioning algorithms perform sub-pixel edge extraction or contour extraction on the original image of the calibration plate, and perform ellipse fitting to extract the image coordinates of the center of the circle. Although the coordinates of the center of the circle can be accurately extracted, there are still the following deficiencies : To operate on the original image of the calibration board image, the algorithm efficiency is low and the running time is long when targeting the high-resolution calibration board image; when performing sub-pixel edge or contour extraction, there are many parameter settings, and the calibration needs to be repeatedly adjusted. For Debuggers have high professional knowledge requirements; high image quality requirements for calibration board images are very sensitive to noise and light effects, and the robustness is not strong

Images