Journal
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
Volume 66, Issue 1-4, Pages 431-443Publisher
SPRINGER LONDON LTD
DOI: 10.1007/s00170-012-4338-2
Keywords
Engineering ceramics; Surface damage; Defect detection; Texture removal; Decision tree
Funding
- National Natural Sciences Foundation of China [51075296]
Ask authors/readers for more resources
As the engineering ceramic ground workpieces usually contain machining damage such as breaks and cracks, the traditional test methods cannot accurately reflect the real surface. Therefore, this paper describes an automatic damage detection system of the engineering ceramic machined surface using image processing techniques, pattern recognition, and machine vision. First, it has great influence on the exact identification of surface damage if engineering ceramic machined surfaces contain grinding texture, so Fourier transform is skillfully adopted to remove grinding texture. Second, through image noise reduction, contrast enhancement, and image segmentation, an optimal combination of image preprocessing is obtained. Then, by comprehensive extraction of surface feature parameters, decision tree classifier based on the C4.5 algorithm is built according to shape features and texture features. Finally, the paper achieves automatic extraction and classification of engineering ceramic grinding surface damage, and the recognition accuracy of breakage reaches over 93 %. Experimental results show that this method is effective in defect detection of the engineering ceramic surface, and it also can provide some analytical basis for the post-function hierarchy partition of engineering ceramic workpieces.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available