Research on intelligent detection methods for defects in surface grinding of steel billets
The surface defect of billet is one of the important factors affecting the strength and toughness of billets,and effective grinding of billet surface is a key step to eliminate quality issues.Traditional rough grinding methods suffer from insufficient and coarse grinding,low efficiency of manually assisted grinding.Therefore,it is urgent to introduce intelligent defect detection technology to accurately locate the locations of surface defects using artificial intelligence methods to guide the fixed-point grinding by a grinding mill,and thereby improving the automation level of grinding.In response to the current problems,this study proposes an intelligent defect detection method for billet surface grinding,which uses a YOLO-based tar-get detection model and designs a defect region aggregation method according to the motion principle of the grinding mill.This ultimately improves the accuracy of defect detection and grinding efficiency,providing advanced intelligent technology support for billet surface grinding.
billet surface defect detectiondefect region aggregationdeep learning
NETL
NSTL
万方数据
