Research On Edge Camber Detection Method Of Hot Rolled Steel Based On Hough Transform

Abstract

The edge camber detection of hot rolled steel mainly depends on human eyes and sensors, but its accuracy is unsatisfactory and there is a certain delay. Therefore, the scheme of using machine vision to detect the bending of hot rolled steel was explored in this paper. Based on the practical engineering problems of hot-rolled plate bending detection, aiming at improving the imaging identification, effectiveness and industrial applicability, an identification method and some key technologies of edge camber detection image were explored in tins paper. Combined with the real hot-rolled steel bending image, the operators of Gaussian, Sobel and Laplace, and Hough transform were used for image preprocessing and line detection respectively. After defining the bending coefficient, the plate bending degree was divided into five grades. And the experiment shows that the machine vision method explored in this paper can accurately judge whether the hot-rolled steel is bent and grade the bending degree.