FEATURE POINT DETECTING METHOD OF WELDING JOINT USING LASER VISION SYSTEM

摘要

The present invention relates to a method of detecting a feature point of a welded joint using a laser vision system, comprising: binarizing an image of a welding object obtained through a laser vision system; Detecting candidate points where feature points may occur by applying an edge extraction algorithm to the binarized image; Detecting a center of the detected candidate points; And comparing the neighboring binarized images with the center of the detected candidate points to determine whether the candidate points are minutiae points. According to the present invention, when detecting minutiae points of a welded joint in a welded joint image measured by a laser vision system, minutiae points of a welded joint can be automatically detected without any additional setting by a user, and even in an image with noise such as spatter or reflected light, Can be detected without any additional computation.