A circular object detection method based on gradient correlation

Abstract

The invention provides a circular target detection method based on gradient correlation, A Sobel edge detection processing is carried out on that original image, the binarization results of image edges are obtained, the vector value of each coordinate point is calculated, the correlation coefficients between the gradient and different coordinate vector values are calculated in turn, The number ofedge points on the circumference of the circle with the center of the circle as the center and different radii is counted, and different radii are traversed from large to small. When the radius is notmore than the corresponding edge points, the radius of the circle target is obtained. The invention determines the position and the radius of the center of the circle by calculating the correlation coefficient of the gradient of each edge point and its relative coordinates, solves the problem of fast detection of the circular target, not only improves the detection accuracy but also reduces the calculation amount, and is widely applicable to the industrial fields such as circular device screening, intelligent transportation and the like.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for detecting a circular target. Background technique [0002] In the field of machine vision, it often involves some detection problems for circular targets, such as: screening of circular devices in industrial production, recognition of circular warning signs in intelligent transportation, etc. Hough transform is one of the most commonly used methods for circle detection because of its strong robustness to image edge discontinuity and noise. However, because the traditional Hough transform needs to traverse the three-dimensional parameter space every time when performing circle detection, the calculation amount is large and the calculation time is long, which is difficult to meet the needs of practical applications. Although some improved methods, such as random Hough transform, can be improved in terms of storage and calculation, random point selection also ...

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the deficiencies of the prior art, the present invention provides a simple and effective circular target detection method, which overcomes the disadvantages of large amount of calculation and low detection accuracy in the traditional method

Images