A digital image corner detection method, system and computer-readable storage medium

Abstract

A method, system and computer-readable storage medium for corner detection of a digital image, the method comprising the steps of: an input image is a binarized two-dimensional image; and an image outer contour of the binarized two-dimensional image is extracted; Perform K traversal on the outer contour of the image to detect candidate corners, K>1; record the candidate corners detected by each traversal; group all the candidate corners; select one of the candidate corners in each group As the corners of the group, the final corners of the image are obtained. The present invention detects and obtains candidate corner points by traversing the pixel points of the outer contour multiple times, and groups and filters the candidate corner points to obtain the final corner position. This method can accurately find all the real corner points in the image contour, and has the characteristics of high positioning accuracy and strong robustness. .

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a distance-based digital image corner detection method, system and computer-readable storage medium. Background technique [0002] The corner point refers to the point with the maximum curvature on the boundary curve in the image, or in other words, the gray value near this point will change drastically. Corner detection is a method used in computer vision systems to obtain image features. It is widely used in motion detection, image matching, video tracking, 3D modeling, and target recognition. It is also called feature point detection. The existing corner detection methods are mainly divided into two categories: one is to use the points on the edge contour of the image to calculate the edge curvature or angle to determine the corner according to the edge contour characteristics of the image; the other is to use the gray information of the image to detect the corner de...

AI Technical Summary

Problems solved by technology

Since the calculation of the curvature of the contour points requires the calculation of the first and second derivatives, if the neighborhood involved in the calculation of the curvature is too small, the corner detection algorithm will be very sensitive to local changes, and the accuracy of the detected corner position is relatively low. Low; and the method of using gray information to locate corners is not very accurate

Images