Abnormal Conductive State Identification of the Copper Rod in a Nickel Electrolysis Procedure Based on Infrared Image Features and Position Characteristics

In the nickel electrolysis industry, detection of conductive state copper rods is an important part production procedure management, as abnormal states rod result in a decline quality electrodeposited plate. Conventional treatment consists manual and handing, which inefficient induces more problems, such safety insulation. Because are only located between busbar, it has obvious position characteristics, also induce calorific difference particular area, can be detected infrared image. We use feature characteristics to identify faults. This paper introduces method practice for identification faults using computer vision theory, including image segmentation with extraction, fault SVM (support vector machine). The shows that divide into heating conditions, normal operating open circuit 90% accuracy rate on obtained samples.