Camber measurement system in a hot rolling mill

Camber is one of the most significant defects in the first stages of hot rolling of steel plates. This kind of defect may cause the clogging of finishing mills, but it is also the visible effect of alterations in the process. In this paper we describe the design and implementation of a computer vision system for real-time measurement of camber in a hot rolling mill. Our goal is to provide valuable feedback information to improve AGC operation. As ground truth values are almost impossible to obtain, we have analyzed the relationship among measured camber and other process variables in order to validate our results. The system has proved to be robust, and at the same time there is a strong relationship between known problems in the mill and system readings.