Multi-ellipse detection and evaluation algorithm and device based on image gradient information, terminal and readable storage medium

Abstract

The invention relates to the technical field of computer vision and image processing, in particular to a multi-ellipse detection and evaluation algorithm and device based on image gradient information, a terminal and a readable storage medium, and the method comprises the steps: S1, extracting feature points from a to-be-tested image through an edge detection algorithm; s2, randomly extracting three points from the feature point set, and substituting the three points into an ellipse evaluation algorithm to evaluate an algebraic coefficient of an ellipse; s3, judging whether the evaluated ellipse is valid or not according to the algebraic coefficient obtained by the evaluated ellipse; s4, searching all inner points in the feature point set; s5, substituting the interior points into an ellipse evaluation algorithm, re-evaluating the algebraic coefficient of the ellipse; S6, judging whether the evaluated ellipse is valid or not according to the evaluated algebraic coefficient of the ellipse and the interior point set; and S7, recording evaluation parameters, and removing interior points. According to the method and the device, the problems of high memory requirement, long calculationtime, inaccurate evaluation result and the like when the traditional Hough transform is used for detecting and evaluating complex geometrical elements are solved.

Description

【Technical field】 [0001] The present application relates to the technical fields of computer vision and image processing, and in particular to a multi-ellipse detection and evaluation algorithm, device, terminal and readable storage medium based on image gradient information. 【Background technique】 [0002] Detecting multiple ellipses contained in the image and evaluating the parameters of each detected ellipse have certain applications in automatic production. Many industrial products contain circular components, and the circle is imaged as an approximate ellipse in the image after the perspective projection transformation of the camera. Detecting and evaluating the ellipse in the image can help to count the number of products in the image, and can also perform geometric matching with the corresponding ellipse features in the product CAD model to determine whether the product has quality problems. [0003] The Hough transform is a traditional algorithm for detecting basic ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of this application is to provide a multi-ellipse detection and evaluation algorithm, device, terminal, and readable storage medium based on image gradient information. Through the algorithm of image gradient information and random sampling consistency, multiple ellipses can be detected from the image. Fast detection and accurate evaluation overcome the problems of large memory requirements, long calculation time, and inaccurate evaluation results when traditional Hough transform detects and evaluates complex geometric elements.

Images