22 May 2017 Vision-based surface defect inspection for thick steel plates
Jong Pil Yun, Dongseob Kim, KyuHwan Kim, Sang Jun Lee, Chang Hyun Park, Sang Woo Kim
Author Affiliations +
Abstract
There are several types of steel products, such as wire rods, cold-rolled coils, hot-rolled coils, thick plates, and electrical sheets. Surface stains on cold-rolled coils are considered defects. However, surface stains on thick plates are not considered defects. A conventional optical structure is composed of a camera and lighting module. A defect inspection system that uses a dual lighting structure to distinguish uneven defects and color changes by surface noise is proposed. In addition, an image processing algorithm that can be used to detect defects is presented in this paper. The algorithm consists of a Gabor filter that detects the switching pattern and employs the binarization method to extract the shape of the defect. The optics module and detection algorithm optimized using a simulator were installed at a real plant, and the experimental results conducted on thick steel plate images obtained from the steel production line show the effectiveness of the proposed method.
© 2017 Society of Photo-Optical Instrumentation Engineers (SPIE) 0091-3286/2017/$25.00 © 2017 SPIE
Jong Pil Yun, Dongseob Kim, KyuHwan Kim, Sang Jun Lee, Chang Hyun Park, and Sang Woo Kim "Vision-based surface defect inspection for thick steel plates," Optical Engineering 56(5), 053108 (22 May 2017). https://doi.org/10.1117/1.OE.56.5.053108
Received: 10 November 2016; Accepted: 26 April 2017; Published: 22 May 2017
Lens.org Logo
CITATIONS
Cited by 37 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Defect inspection

Light sources and illumination

Cameras

Computer simulations

Defect detection

Detection and tracking algorithms

Image processing

RELATED CONTENT

Aircraft recognition and tracking device
Proceedings of SPIE (September 28 2011)

Back to Top