An online inspection system of surface defects for copper strip based on computer vision

In this paper, an online system based on computer vision has been designed using modularized frame to inspect surface defects in copper strips which affect the quality of copper strips. First, a prejudgment algorithm is advocated to identify whether the copper strip's surface images captured by linear charged couple device camera have defects. Then multiple features such as invariant moment, breadth-length ratio, circular degree and rectangular degree are extracted as feature vector of defect. Then the type of defect is identified by multiple features fuzzy classifier and warning signals like light, sound, and so on, are given according to different defect ranks. Finally, defect information and video of copper strip are both stored into storage. The system is put into practice in a factory of copper strip and is also proved to be applicable and effective.