Vision-Based Camber Measurement System in Hot Rolling Process

Abstract

In hot rolling process, camber is one of the most significant defects where camber is flatness asymmetries produced by non-uniform distribution of rolling pressure.This kind of defect may cause not only the clogging of finishing mill, but also visible defects such as strip edge folds, edge cracks, holes, and scrapes in the process.In order to reduce camber for hot rolled steel strips, it is important to measure the exact camber value with information of the overall shape. This paper describes a computer vision system with new algorithm for obtaining the optimal cutting line, overall strip shape, and camber value from two area scan Charge Coupled Cameras (CCD) in hot rolling process. The proposed algorithm consists of three parts: cutting line detection, image stitching, and camber detection. The cutting line detection part determines the optimal cutting line of the head and tail part of strip for endless hot rolling process. The image stitching part obtains the overall shape of a hot rolled steel strip with continuous image sequence obtained by two area scan cameras. The camber detection part measures head camber, tail camber, and whole camber of the steel strip. Through the proposed algorithm, the system satisfies time constraints for detecting the camber of the steel strip.