Abstract
Steel ball's surface detection plays a significant role in computer vision applications. An efficient and accurate defect detection approach is implemented in this paper. The defect region on steel ball's surface is achieved by region growing technique on steel ball's surface images. The defect region such as spot, cluster spots, scratch and strip defect is classified by kernel extreme learning machine. The experimental results are carried out by using steel ball's surface images and perfect performance is achieved by the proposed defect detection technique.
Full Text
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call