An automatic surface inspection system for flat rolled steel

The paper describes a feasibility study of a system to classify visible defects in flat rolled steel on-line, as the steel emerges from the rolling mill. A visual signal is developed by a television camera, and this signal is processed by a hardware system which stores information from the scans and alerts a small computer when a possible flaw has been detected. Executing a pattern recognition algorithm, the computer makes the final detection and classification decision before the next coil is rolled.