Selection of significant visual features for classification of scales using boosting trees model

Szymon Lechwar

ArcelorMittal Poland, Hot Rolling Mill in Kraków, ul. Ujastek 1, 30-969 Kraków.

DOI:

https://doi.org/10.7494/cmms.2013.2.0444

Abstract:

The subject of this paper is to design and implement an efficient model for various kinds of scales recognition at the Hot Rolling Mill (HRM) in Kraków. Subsequently, the model and its most important variables can be used to describe and distinguish different kinds of scales. At the moment an extensive knowledge regarding the reasons of scale occurrence is gathered. Nevertheless, the real challenges nowadays seem to be measuring techniques of those phenomena, as well as reliable online classification. This paper describes the basics of automatic surface inspection system (ASIS) which was used as a source of entry data, as well as the method of interpretation of the data obtained from this system. The ASIS provided numerous features describing single image, which was considered as a defect. The objective of this paper was to supply information regarding the most important visual attributes, which will be subsequently used in building reliable classifier for scale recognition. It was done by use of data mining techniques. The result was a set of measurement data, stored in online production database. However, some kinds of scales could not be recognized efficiently. The reason behind that was the lack of unique features, which could distinguish them from the other defects. This problem will be solved in following studies by creating offline post processing rules.

Cite as:

Lechwar, S. (2013). Selection of significant visual features for classification of scales using boosting trees model. Computer Methods in Materials Science, 13(2), 289 – 294. https://doi.org/10.7494/cmms.2013.2.0444

Article (PDF):