Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization
Accurate detection of defects on steel surfaces is crucial for maintaining quality standards in steel production. This paper addresses the challenge of classifying steel sheets into distinct defect categories by presenting a robust method that leverages deep learning and advanced optimization techni...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Systems and Soft Computing |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772941924000991 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850053632105381888 |
|---|---|
| author | Prabira Kumar Sethy Laxminarayana Korada Santi Kumari Behera Akshay Shirole Rajat Amat Aziz Nanthaamornphong |
| author_facet | Prabira Kumar Sethy Laxminarayana Korada Santi Kumari Behera Akshay Shirole Rajat Amat Aziz Nanthaamornphong |
| author_sort | Prabira Kumar Sethy |
| collection | DOAJ |
| description | Accurate detection of defects on steel surfaces is crucial for maintaining quality standards in steel production. This paper addresses the challenge of classifying steel sheets into distinct defect categories by presenting a robust method that leverages deep learning and advanced optimization techniques. We propose a novel approach that utilizes the ResNet101 model to extract deep features, which are then classified using a support vector machine (SVM). To enhance the SVM's performance, Bayesian optimization is employed for hyperparameter tuning. Our method is validated using the ''Severstal: Steel Defect Detection'' dataset from Kaggle, achieving a validation accuracy of 89.1 % and a test accuracy of 90.6 %, with a classification error of 0.10934. Additionally, the area under the curve (AUC) for each class exceeds 0.95 in both the validation and test sets, demonstrating excellent discriminatory power. Further evaluation on the DAGM dataset achieved flawless results, with an accuracy of 100 %, AUC of 1, sensitivity of 100 %, specificity of 100 %, precision of 100 %, MCC of 100 %, F1 score of 1, and kappa of 100 %. On the NEU dataset, our method achieved an accuracy of 97.92 %, sensitivity of 97.92 %, specificity of 99.58 %, precision of 98.06 %, F1 score of 0.9791, MCC of 97.55 %, and kappa of 92.50 %. These results demonstrate the robustness and adaptability of the proposed method, offering an efficient and reliable solution for automating steel defect detection and surface defect classification in industrial applications. |
| format | Article |
| id | doaj-art-c6ac80ea7791461780ae1f2c35f6a161 |
| institution | DOAJ |
| issn | 2772-9419 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Systems and Soft Computing |
| spelling | doaj-art-c6ac80ea7791461780ae1f2c35f6a1612025-08-20T02:52:28ZengElsevierSystems and Soft Computing2772-94192024-12-01620017010.1016/j.sasc.2024.200170Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimizationPrabira Kumar Sethy0Laxminarayana Korada1Santi Kumari Behera2Akshay Shirole3Rajat Amat4Aziz Nanthaamornphong5Department of Electronics and Communication Engineering, Guru Ghasidas Vishwavidyalaya, Bilaspur, C.G., 495009, IndiaMicrosoft, Redmond, Washington, United StatesDepartment of Computer Science and Engineering, Veer Surendra Sai University of Technology, Sambalpur, Odisha, 768018, IndiaDr. C.V.Raman University, Bilaspur, Chhattisgarh, IndiaDepartment of Electronics and Communication Engineering, Indian Institute of Technology, Guwahati, Assam, 781039, IndiaCollege of Computing, Prince of Songkla University, Phuket, Thailand; Corresponding author.Accurate detection of defects on steel surfaces is crucial for maintaining quality standards in steel production. This paper addresses the challenge of classifying steel sheets into distinct defect categories by presenting a robust method that leverages deep learning and advanced optimization techniques. We propose a novel approach that utilizes the ResNet101 model to extract deep features, which are then classified using a support vector machine (SVM). To enhance the SVM's performance, Bayesian optimization is employed for hyperparameter tuning. Our method is validated using the ''Severstal: Steel Defect Detection'' dataset from Kaggle, achieving a validation accuracy of 89.1 % and a test accuracy of 90.6 %, with a classification error of 0.10934. Additionally, the area under the curve (AUC) for each class exceeds 0.95 in both the validation and test sets, demonstrating excellent discriminatory power. Further evaluation on the DAGM dataset achieved flawless results, with an accuracy of 100 %, AUC of 1, sensitivity of 100 %, specificity of 100 %, precision of 100 %, MCC of 100 %, F1 score of 1, and kappa of 100 %. On the NEU dataset, our method achieved an accuracy of 97.92 %, sensitivity of 97.92 %, specificity of 99.58 %, precision of 98.06 %, F1 score of 0.9791, MCC of 97.55 %, and kappa of 92.50 %. These results demonstrate the robustness and adaptability of the proposed method, offering an efficient and reliable solution for automating steel defect detection and surface defect classification in industrial applications.http://www.sciencedirect.com/science/article/pii/S2772941924000991Steel sliceDefect detectionDeep featureSupport vector machineResNet101Bayesian optimization |
| spellingShingle | Prabira Kumar Sethy Laxminarayana Korada Santi Kumari Behera Akshay Shirole Rajat Amat Aziz Nanthaamornphong Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization Systems and Soft Computing Steel slice Defect detection Deep feature Support vector machine ResNet101 Bayesian optimization |
| title | Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization |
| title_full | Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization |
| title_fullStr | Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization |
| title_full_unstemmed | Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization |
| title_short | Maximizing steel slice defect detection: Integrating ResNet101 deep features with SVM via Bayesian optimization |
| title_sort | maximizing steel slice defect detection integrating resnet101 deep features with svm via bayesian optimization |
| topic | Steel slice Defect detection Deep feature Support vector machine ResNet101 Bayesian optimization |
| url | http://www.sciencedirect.com/science/article/pii/S2772941924000991 |
| work_keys_str_mv | AT prabirakumarsethy maximizingsteelslicedefectdetectionintegratingresnet101deepfeatureswithsvmviabayesianoptimization AT laxminarayanakorada maximizingsteelslicedefectdetectionintegratingresnet101deepfeatureswithsvmviabayesianoptimization AT santikumaribehera maximizingsteelslicedefectdetectionintegratingresnet101deepfeatureswithsvmviabayesianoptimization AT akshayshirole maximizingsteelslicedefectdetectionintegratingresnet101deepfeatureswithsvmviabayesianoptimization AT rajatamat maximizingsteelslicedefectdetectionintegratingresnet101deepfeatureswithsvmviabayesianoptimization AT aziznanthaamornphong maximizingsteelslicedefectdetectionintegratingresnet101deepfeatureswithsvmviabayesianoptimization |