Change search
CiteExportLink to record
Permanent link

Direct link
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Classification of Carbide Distributions using Scale Selection and Directional Distributions
KTH, Superseded Departments, Numerical Analysis and Computer Science, NADA.ORCID iD: 0000-0002-9081-2170
1997 (English)In: Proc. 4th International Conference on Image Processing: ICIP'97, 1997, Vol. II, 122-125 p.Conference paper, Published paper (Refereed)
Abstract [en]

This paper presents an automatic system for steel quality assessment, by measuring textural properties of carbide distributions. In current steel inspection, specially etched and polished steel specimen surfaces are classified manually under a light microscope, by comparisons with a standard chart. This procedure is basically two-dimensional, reflecting the size of the carbide agglomerations and their directional distribution. To capture these textural properties in terms of image features, we first apply a rich set of image-processing operations, including mathematical morphology, multi-channel Gabor filtering, and the computation of texture measures with automatic scale selection in linear scale-space. Then, a feature selector is applied to a 40-dimensional feature space, and a classification scheme is defined, which on a sample set of more than 400 images has classification performance values comparable to those of human metallographers. Finally, a fully automatic inspection system is designed, which actively selects the most salient carbide structure on the specimen surface for subsequent classification. The feasibility of the overall approach for future use in the production process is demonstrated by a prototype system. It is also shown how the presented classification scheme allows for the definition of a new reference chart in terms of quantitative measures.

Place, publisher, year, edition, pages
1997. Vol. II, 122-125 p.
Machine Vision and Applications, ISSN 0932-8092 ; 12
National Category
Computer Science Computer Vision and Robotics (Autonomous Systems) Mathematics
URN: urn:nbn:se:kth:diva-40237DOI: 10.1109/ICIP.1997.632012ISI: 000089927000002OAI: diva2:440708
4th International Conference on Image Processing

QC 20110913

Available from: 2013-04-05 Created: 2011-09-13 Last updated: 2013-04-05Bibliographically approved

Open Access in DiVA

fulltext(1157 kB)