Detection of small surface defects using DCT based enhancement approach in machine vision systems

Fuqiang He; Wen Wang; Zichen Chen

doi:10.1117/12.664153

21 March 2006 Detection of small surface defects using DCT based enhancement approach in machine vision systems

Fuqiang He, Wen Wang, Zichen Chen

Proceedings Volume 6040, ICMIT 2005: Mechatronics, MEMS, and Smart Materials; 60400H (2006) https://doi.org/10.1117/12.664153
Event: ICMIT 2005: Merchatronics, MEMS, and Smart Materials, 2005, Chongqing, China

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%.

Citation Download Citation

Fuqiang He, Wen Wang, and Zichen Chen "Detection of small surface defects using DCT based enhancement approach in machine vision systems", Proc. SPIE 6040, ICMIT 2005: Mechatronics, MEMS, and Smart Materials, 60400H (21 March 2006); https://doi.org/10.1117/12.664153

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available