A surface defect detection method for rolling magnesium alloy sheet based on computer vision

In the rolling process of magnesium alloy sheet, defects such as edge crack, fold and ripple are easy to appear on the surface of the sheet. These defects will affect the product yield and quality, and cause waste of resources. In this paper, computer vision technology is used to analyze the image of rolling magnesium alloy sheet in real-time, extract its defect features, and Bayesian classifier and Random Forest (RF) classifier are used to identify defects. The experimental results show that the comprehensive defect recognition rate of the RF algorithm is up to 92,4 %, which is much higher than the accuracy of Bayesian classifier, and it is more suitable for the recognition of surface defects of magnesium sheet