Camber Detection Algorithm using the Image Stitching Technique in Hot-Rolling Process

J.W. Yoo, N.W. Kong, J.-m. Song, and P. Park (Korea)


camber detection, hot-rolling process, image stitching, computer vision


Camber is one of the most significant defects in hot-rolling process, where camber is longitudinal curvature in the plan view of a slab or strip. The information of the overall shape is important to determine exact camber value. Therefore, this paper proposes new algorithm for obtaining the overall strip shape and camber value only using the image data. The proposed algorithm consists of two parts: image stitching to obtain the overall shape of a hot-rolled strip and camber detection to obtain camber of the strip. The image stitching part combines continuous image sequence obtained by two area scan cameras to make the overall image of the strip with the proposed image stitching technique. The camber detection part calculates the camber of the overall image. Through the above parts, the proposed algorithm not only stitches the continuous strip image sequence but also determines the camber.

Important Links:

Go Back