Image contour corner detection method based on centroid distance calculation

Abstract

The invention discloses an image contour corner detection method based on centroid distance calculation. An Euclidean distance positioning angular point area of each image contour point and a local area centroid formed by the image contour point is calculated; the method comprises the following specific steps: firstly, an edge contour is extracted, and the Euclidean distance between each pixel point of the image contour and the center of mass of a local region formed by the pixel point is calculated so that a distance set G is formed, upper and lower limit threshold processing and point numberthreshold processing are conducted on the G to form a set I, finally, a Harris angular corner detector is applied to each point in each element in the set I to calculate a response, and the coordinate of the corresponding point with the strongest response in each element is the angular corner. According to the invention, the Euclidean distance between each image contour point and the centroid ofthe local area formed by the image contour point is calculated to locate the angular point area; the calculation result is slightly influenced by the change of the image gray scale, the immunity to noise is enhanced, the generation of pseudo corners is reduced, and the method is suitable for the computer vision fields of image registration and matching, image fusion, target recognition and the like.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an image contour corner point detection method based on centroid distance calculation. Background technique [0002] Image corners are strong feature points of images, which are widely used in object recognition, image registration, motion detection and other fields. Existing corner detection methods include: corner detection based on grayscale images and corner detection based on binary images, wherein the Harris corner detection method based on grayscale images has the advantages of two-dimensional translation, rotation, and a small amount of illumination. Changes, a small amount of viewing angle changes are not sensitive, and the advantages of a small amount of calculation are widely used. [0003] The traditional Harris corner detection method is based on the gray value matrix of the block area formed by the pixel point as the center. When the illuminati...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the traditional Harris corner detection method, which is sensitive to noise when performing corner detection, the detection result is greatly affected by the change of image gray scale, and there are many false corners, and provides an image contour angle based on centroid distance calculation. Point detection method, this method can greatly reduce the impact of image gray level changes on corner positioning, avoid the weakening or loss of corner points, and reduce the generation of false corner points

Images