Share Email Print

Proceedings Paper

Detection of small surface defects using DCT based enhancement approach in machine vision systems
Author(s): Fuqiang He; Wen Wang; Zichen Chen
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Utilizing DCT based enhancement approach, an improved small defect detection algorithm for real-time leather surface inspection was developed. A two-stage decomposition procedure was proposed to extract an odd-odd frequency matrix after a digital image has been transformed to DCT domain. Then, the reverse cumulative sum algorithm was proposed to detect the transition points of the gentle curves plotted from the odd-odd frequency matrix. The best radius of the cutting sector was computed in terms of the transition points and the high-pass filtering operation was implemented. The filtered image was then inversed and transformed back to the spatial domain. Finally, the restored image was segmented by an entropy method and some defect features are calculated. Experimental results show the proposed small defect detection method can reach the small defect detection rate by 94%.

Paper Details

Date Published: 21 March 2006
PDF: 6 pages
Proc. SPIE 6040, ICMIT 2005: Mechatronics, MEMS, and Smart Materials, 60400H (21 March 2006); doi: 10.1117/12.664153
Show Author Affiliations
Fuqiang He, Zhejiang Univ. (China)
Wen Wang, Zhejiang Univ. (China)
Zichen Chen, Zhejiang Univ. (China)

Published in SPIE Proceedings Vol. 6040:
ICMIT 2005: Mechatronics, MEMS, and Smart Materials
Yunlong Wei; Kil To Chong; Takayuki Takahashi; Shengping Liu; Zushu Li; Zhongwei Jiang; Jin Young Choi, Editor(s)

© SPIE. Terms of Use
Back to Top