Use of Artificial Intelligence in Classification of Mill Scale Defects
Abstract
The subject of the work was to design and implement classification model for various kinds of mill scales recognition at typical hot rolling mill (HRM). Input data were measured by the automatic surface inspection system, which provided numerous features describing single image, considered as a defect. The data were analyzed by using feature selection methods. Afterwards, determined most important subsets of features, were applied to build scale classification models. The most efficient model and its key variables were used to distinguish different kinds of scales. Implementation of the model in production system allowed to show a gain of overall mill scales classification accuracy. In the paper, the main attention is put on the research methodology and artificial intelligence (AI) methods used to classify scale defects.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Decision trees, random forests, and relevance vector machines (RVMs) have achieved good effects under supervised learning for flat steel surface defect classification.
“In addition to the above classifiers, some other classifiers have also achieved good effects under supervised learning for the task of flat steel surface defect classification, such as decision trees, random forests, and relevance vector machines (RVMs).”
From Automated Visual Defect Classification for Flat Steel Surface: A Survey (2020) · cites this paper as [110]
Cites
A logical calculus of the ideas immanent in nervous activity
McCulloch, W. S., & Pitts, W. (1943) · The Bulletin of Mathematical Biophysics · doi:10.1007/bf02478259 · Cited by 18,566 · see citing works
On the mean accuracy of statistical pattern recognizers
Hughes, G. W. (1968) · IEEE Transactions on Information Theory · doi:10.1109/tit.1968.1054102 · Cited by 2,890 · see citing works
Thurstone, L. L. (1931) · Psychological Review · doi:10.1037/h0069792 · Cited by 581 · see citing works
39 Dimensionality and sample size considerations in pattern recognition practice
Jain, A. K., & Chandrasekaran, B. (1982) · Handbook of statistics · doi:10.1016/s0169-7161(82)02042-2 · Cited by 560 · see citing works
Ginzburg (1989) · CRC Press eBooks · doi:10.1201/9781466593510 · Cited by 224 · see citing works
A Comparison of Seven Techniques for Choosing Subsets of Pattern Recognition Properties
Mucciardi, A. N., & Gose, E. E. (1971) · IEEE Transactions on Computers · doi:10.1109/t-c.1971.223398 · Cited by 203 · see citing works
Agarwal, K., Shivpuri, R., Zhu, Y., Chang, T.-S., & Huang, H. (2011) · Expert Systems with Applications · doi:10.1016/j.eswa.2010.12.026 · Cited by 54 · see citing works
Surface characteristics of oxide scale in hot strip rolling
Sun, W., Tieu, A. K., Jiang, Z., Lü, C., & Zhu, H. (2003) · Journal of Materials Processing Technology · doi:10.1016/s0924-0136(03)00823-9 · Cited by 49 · see citing works
(2000) · doi:10.1109/kes.2000.885769 · Cited by 45 · see citing works
An automatic surface inspection system for flat rolled steel
Saridis, G. N., & Brandin, D. M. (1979) · Automatica · doi:10.1016/0005-1098(79)90001-3 · Cited by 19 · see citing works
Effect of steel composition on failure of oxide scales in tension under hot rolling conditions
Tan, K. S., Krzyżanowski, M., & Beynon, J. H. (2001) · Steel Research · doi:10.1002/srin.200100113 · Cited by 17 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Automated Visual Defect Classification for Flat Steel Surface: A Survey
Luo, Q., Fang, X., Su, J., Zhou, J., Zhou, B., Yang, C., Liu, L., Gui, W., & Tian, L. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3030167 · Cited by 148 · see citing works
Multiobjective Optimization of Roll-Forming Procedure Using NSGA-II and Type-2 Fuzzy Neural Network
Qazani, M. R. C., Bidabadi, B. S., Asadi, H., Nahavandi, S., & Bidabadi, F. S. (2023) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2023.3287921 · Cited by 24 · see citing works
Artificial intelligence in healthcare services: past, present and future research directions
Shah, W. S., Elkhwesky, Z., Jasim, K. M., Elkhwesky, E. F. Y., & Elkhwesky, F. F. Y. (2023) · Review of Managerial Science · doi:10.1007/s11846-023-00699-w · Cited by 28 · see citing works
Steel surface defects analysis with machine vision and deep learning
Frydrych, K., Tomczak, M., Jasinski, J. J., & Papanikolaou, S. (2025) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-025-16539-y · Cited by 10 · see citing works
Zhou, Y., Mao, Y., Xuan, D., Jiang, T., Fan, W., Zhu, B., & Zhou, C. P. (2023) · steel research international · doi:10.1002/srin.202200953 · Cited by 7 · see citing works
Scale Formation on 430 Stainless Steel in a Simulated Slab Combustion Reheat Furnace Atmosphere
Osei, R., Lekakh, S. N., & O’Malley, R. J. (2020) · Iron & Steel Technology Conference and Exposition (AISTech) · doi:10.33313/380/120 · Cited by 3 · see citing works
Comparison Of Iron And Steel Production Defects Using Classification Algorithms
Akinci, I. B., Alobaidi, D., & Ersöz, F. (2021) · International Congress on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA) · doi:10.1109/hora52670.2021.9461299 · Cited by 2 · see citing works
Descaling of Medium C and High Si, Mn Steels
Osei, R., Lekakh, S., O’Malley, R. J., Peterson, L., & Sasso, O. (2021) · Iron & Steel Technology Conference and Exposition (AISTech) · doi:10.33313/382/060 · Cited by 2 · see citing works
Prediction of Waviness Values in Skew Rolling Using Machine Learning Methods
Lis, K., Pater, Z., Tomczak, J., Bulzak, T. A., & Kusiak, T. (2023) · Advances in Science and Technology – Research Journal · doi:10.12913/22998624/172338 · Cited by 2 · see citing works
Lis, K. (2023) · Materials · doi:10.3390/ma16227136 · Cited by 1 · see citing works
Banerjee, A., Kedarnath, T., Sadhukhan, P., Ghosh, K. S., & Ghosh, M. M. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.115314
- Authors
- Lechwar, S., Rauch, Ł., & Pietrzyk, M.
- Affiliations
- ArcelorMittal (Poland) (PL)
- AGH University of Krakow (PL)
- Corresponding author
- Lechwar, Szymon
- Venue
- steel research international
- Publisher
- Wiley
- Published
- 2014-06-16
- Volume
- 86
- Issue
- 3
- Pages
- 266-277
- ISSN
- 1611-3683 (ISSN-L)
- 1869-344X
- OpenAlex
- W2012454811
- Cited by
- 11
- References
- 11
- Citations by year
- 2026: 1 · 2025: 1 · 2023: 5 · 2021: 2 · 2020: 2
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- steel research international (publishedVersion)
- Topics
- Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Industrial Vision Systems and Defect Detection
- Physical Sciences › Engineering › Mechanical Engineering › Non-Destructive Testing Techniques
- Physical Sciences › Engineering › Civil and Structural Engineering › Infrastructure Maintenance and Monitoring
- Keywords
- hot rolling mill, scale classification, automatic surface inspection, feature selection, steel surface defects, classification models
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
