Visual surface inspection of scale-covered steel billets

Authors:
Jong Pil Yun;SungHoo Choi;Boyeul Seo;Sang Woo Kim
Affiliations:
Pohang University of Science and Technology (POSTECH), Hyojadong, Namgu, Pohang, Korea;Pohang University of Science and Technology (POSTECH), Hyojadong, Namgu, Pohang, Korea;Pohang University of Science and Technology (POSTECH), Hyojadong, Namgu, Pohang, Korea;Pohang University of Science and Technology (POSTECH), Hyojadong, Namgu, Pohang, Korea
Venue:
CGIM '08 Proceedings of the Tenth IASTED International Conference on Computer Graphics and Imaging
Year:
2008

Citing 3
Cited 0

A Two Stage Defect Recognition Method for Parquet Slab Grading

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 4
An Intelligent Real-time Vision System for Surface Defect Detection

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
Visual inspection of machined metallic high-precision surfaces

EURASIP Journal on Applied Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper a visual surface inspection method for scalecovered steel billets is introduced. The billet surface image has a lot of scale, which is oxidized substance unavoidably caused during manufacturing process. Because scale of billet surface is similar to real defects such as crack, it is difficult to distinguish defects from scale-covered image. Moreover, different reflection properties among the defect-free image due to scale on the surface make hard to detect defects. To solve above problems, this paper proposes a new approach for defect detection based on Gabor filters with optimized energy separation. To enhance the performance of detection, double threshold method and Laplacian operation are proposed. Finally, the experimental results conducted on billet surface images obtained real steel making plant.