Detection of surface defects on steel balls using image processing technology

Jian-min Zhou; Yang Yang

doi:10.1117/12.819620

31 December 2008 Detection of surface defects on steel balls using image processing technology

Jian-min Zhou, Yang Yang

Proceedings Volume 7130, Fourth International Symposium on Precision Mechanical Measurements; 713028 (2008) https://doi.org/10.1117/12.819620
Event: Fourth International Symposium on Precision Mechanical Measurements, 2008, Anhui, China

Abstract

Manual methods of surface defect detection and recognition on steel balls which have large workload and poor reliability are widely used in domestic manufactories. The defects of steel balls' surface have several primary categories, including pitting, scuffing, scratch and nicks. An image processing technology for automatic detection of surface defects on steel balls is proposed. The first step is image segmentation. According to defects' images which have different gray-level histogram, iterative method and mode method are adopted to make binarization. Then, a connected component labeling algorithm to sign the connected region in binary image is also presented. The following and a crucial step was feature extraction. General geometry features and moment invariant features of every connected region are calculated for recognition. Eventually, BP neural network is an efficient approach to recognize classification. Experiment show that mainly 95 percent of the surface defect categories have been classified correctly.

Citation Download Citation

Jian-min Zhou and Yang Yang "Detection of surface defects on steel balls using image processing technology", Proc. SPIE 7130, Fourth International Symposium on Precision Mechanical Measurements, 713028 (31 December 2008); https://doi.org/10.1117/12.819620

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