A method for identifying and locating straight lines, ellipses and intersections for uncorrected images

Abstract

The invention discloses a method for identifying and locating straight lines, ellipses and intersections of uncorrected images. The invention includes: inputting a series of line segments, judging whether they belong to the same straight line by using the relationship between the line segments, and dividing them into several line segment sets; splitting the bifurcated lines that were mistakenly merged based on ransac random sampling consistent detection algorithm, and updating the line segment set; Multiplication preliminarily judges the type of line segment set, which is divided into two types: circular line and straight line, and uses mathematical formula to fit and express, unclassified line segment set uses parabolic formula to fit; judges the type of unclassified line segment set again, and corrects Ellipse mathematical formula; for the set of line segments of the straight line class, the intersection point is obtained through the relationship of the fitted mathematical expression formula, and the authenticity of the intersection point is verified; the position information of the straight line, ellipse, and intersection point is finally output. In this way, the complexity is reduced and the degree of linear fitting is improved. Through the above method, the recognition and positioning of straight lines, ellipses and intersections can be realized only through uncorrected images.

Description

technical field [0001] The invention belongs to the field of computer vision, and in particular relates to a method for identifying and locating straight lines, ellipses and intersections of uncorrected images. Background technique [0002] In recent years, intelligent robots are the key research objects, and intelligent robot technology has also developed rapidly. Competitions related to intelligent robots are also in full swing, attracting everyone's attention, and providing a communication platform for scientific researchers engaged in robotics from all over the world. Robocup is the Robot World Cup competition and an important international event in the field of robotics. The robot needs to determine the next movement by identifying its own location and the scene in the field of vision, so the robot's vision system is very important and indispensable. [0003] In the Robocup standard platform competition venue, the white line of the venue is an important object for rob...

AI Technical Summary

Problems solved by technology

However, the disadvantage of this kind of method is that it takes up a certain amount of memory space, and the calculation time is too long, which is difficult to meet the real-time requirements of the algorithm in the football robot field.

In addition, the input image of this type of method requires an undistorted image. For an image with excessive distortion, it is difficult to detect the deformed straight line.

Images