Detection method of checkboard grid image angular point sub pixel

Abstract

The invention provides a detection method of a checkboard grid image angular point sub pixel. The detection method of the checkboard grid image angular point sub pixel comprises the steps that 1, a checkboard grid calibration template is filmed trough a camera, a template image of the checkboard grid calibration template is obtained, and binaryzation is conducted on the template image to obtain a binaryzation image; 2, traversal is conducted on the binaryzation image through a circular template, and screening is conducted on all the tested pixel points of the binaryzation image to screen out candidate angular points; 3, according to the character of the candidate angular points that the candidate angular points are symmetrically arranged near a true angular point which is determined by an angular point set, the abscissa average value and the ordinate average value of the pixel positions of the candidate angular points are calculated, and the abscissa average value and the ordinate average value serve as the sub pixel coordinates of the true angular point. According to the detection method of the checkboard grid image angular point sub pixel, only simple arithmetic is needed, and operations of differential, interpolation, surface fitting and the like are not needed, so that the detection process is simple and easy to realize, sub pixel level precision can be directly obtained, and good detection precision and real-time are obtained.

Description

technical field [0001] The invention belongs to a detection method in the technical field of camera calibration, in particular to a detection method of a checkerboard image corner sub-pixel. Background technique [0002] The checkerboard calibration template is widely used in camera calibration technology because of its simple features, obvious contrast, and easy identification. The accuracy of checkerboard image corner detection directly determines the calibration accuracy of the camera, which in turn affects the accuracy of subsequent tasks such as visual measurement. Therefore, improving the accuracy of corner detection in checkerboard images is an important research topic in the field of visual measurement. Currently, checkerboard image corner detection methods can be divided into corner detection methods based on edge detection and corner detection methods based on grayscale changes. [0003] Prior Art Document 1 and Prior Art Document 2 provide two different methods ...

AI Technical Summary

Problems solved by technology

This type of algorithm avoids the above-mentioned defects in the method of corner detection based on edge detection, and has good comprehensive performance, but there are still some limitations. For example, in the method disclosed in prior art document 3, when in a complex Under background conditions, it is difficult to obtain better detection accuracy. In the methods disclosed in prior art document 4, prior art document 5, and prior art document 6, the accuracy of locating corner points is only at the pixel level. To obtain sub-pixel-level accuracy, follow-up processing is required. Although the method disclosed in prior art document 7 has high detection accuracy, the real-time performance is poor. Although the method disclosed in prior art document 8 has good real-time performance, the exposure degree is poor. The detection accuracy is low when it is large

Images