A hub contour detection method based on improved Hough transform

Abstract

The invention discloses a hub contour detection method based on improved Hough transform, which comprises the following steps: firstly, preprocessing the wheel hub image to extract the edge contour curve of the wheel hub; then dividing the contour boundary points into different subsets and removing the noise set; obtaining the intersection points of each section line and contour line by cutting the images at intervals of 30 degrees, and obtaining the equations of fitting straight lines in different directions by Hough transform; determining the coordinates of the center of the circle, obtaining the distance between the center of the circle and the intersection point, judging whether the circle is circular or not according to its variance, and finally obtaining the center and radius of thecircle. Compared with conventional Hough transform and a random Hough transform algorithm, the detection method of the invention has the advantages of short running time, fast running speed and higherdetection precision.

Description

technical field [0001] The invention belongs to the fields of digital image processing and pattern recognition, and in particular relates to a wheel hub contour detection method based on improved Hough transformation. Background technique [0002] Compared with other types of images, the key feature of the hub image is its outline circle feature. The diameter of the outline circle and the coordinate position of the center of the circle are important parameters of the hub. Identifying the key outline circle features of the hub will facilitate further image analysis and processing. At present, the contour circle feature extraction methods mainly include Hough circle transform method, edge chain code tracking detection method and least square fitting method. Among them, because the Hough transform has good noise robustness, discontinuity adaptability and rotation invariance, the Hough circle transform method is the most effective and commonly used. [0003] The Hough transform...

AI Technical Summary

Problems solved by technology

When the detected image is not very complicated, the traditional Hough transform can get satisfactory results, but because the traditional Hough transform (CHT) is a one-to-many mapping matching algorithm, each feature point (edge ​​pixel) in the image point) matches the parameters of all curves that may pass through the point in the parameter space, and accumulates votes for these parameter units, which makes this method have great defects: large amount of calculation, long calculation time; need to divide the parameter space unit Pre-stored in the storage unit, it consumes a lot of storage space; due to the discretization of image space and parameter space, the detection accuracy of Hough transform is not high; it is impossible to distinguish noise points from feature points that make up the target, etc.

[0004] On this basis, Xu et al. proposed the random Hough transform (RHT) method, by randomly selecting three non-collinear boundary points to map to a point in the parameter space, avoiding the one-to-many mapping in the CHT, but this This method is greatly affected by noise and image complexity, and it is easy to generate a large number of invalid point accumulation, thus affecting the calculation accuracy

Images