Automated On-Line Fast Detection for Surface Defect of Steel Strip Based on Multivariate Discriminant Function

Abstract

Surface inspection of steel strips is of great importance to improve the quality because it is mainly affected by the defects on the surface. Digital image processing methods have been developed for defect detection for past few years. As to an automated on-line detection system, the research on rapid defect detection is quite significant. In this paper, an approach to detect surface defects of steel strip based on multivariate discriminant function is discussed. By subdividing the images into blocks and extracting related features, tiny defects are effectively detected. With the inspection of the defects, a multivariate discriminant function model has been established. Persuasive experiments results were obtained which prove the feasibility and accuracy of the proposed method. Thus, this research is quite practical and lays a solid foundation for the future study.