Surface Defect Detection of Aluminum Plates Using Improved Faster R-CNN and ResNet-50

Xiaomei Ni; David Chua Sing Ngie; Wanzhen Wang; Miaomiao Xin; Na Li; Xiaole Han; Jialei Shi

doi:10.46604/peti.2025.14884

Authors

Xiaomei Ni Faculty of Engineering, Universiti Malaysia Sarawak (UNIMAS), Sarawak, Malaysia
David Chua Sing Ngie Faculty of Engineering, Universiti Malaysia Sarawak (UNIMAS), Sarawak, Malaysia
Wanzhen Wang School of Intelligent Manufacturing and Control Engineering, Qilu Institute of Technology, Jinan, China
Miaomiao Xin School of Intelligent Manufacturing and Control Engineering, Qilu Institute of Technology, Jinan, China
Na Li School of Intelligent Manufacturing and Control Engineering, Qilu Institute of Technology, Jinan, China
Xiaole Han School of Intelligent Manufacturing and Control Engineering, Qilu Institute of Technology, Jinan, China
Jialei Shi School of Intelligent Manufacturing and Control Engineering, Qilu Institute of Technology, Jinan, China

DOI:

https://doi.org/10.46604/peti.2025.14884

Keywords:

deep learning, Faster R-CNN, feature fusion, ResNet-50, surface defect detection

Abstract

This study aims to enhance the accuracy of surface defect detection in aluminum profiles. To address low recognition accuracy caused by irregular defect sizes and the coexistence of multiple defects, an inspection system integrating Faster Region-based Convolutional Neural Network (Faster R-CNN) and Residual Networks (ResNet-50) is proposed. After data enhancement and preprocessing of the aluminum profile image dataset from the Tianchi platform, ResNet-50 is used to extract deep features, and the Region Proposal Network (RPN) within Faster R-CNN is applied to generate candidate regions for classification and localization. Experimental evaluations demonstrate that the proposed model identifies all defect types with over 93% accuracy, while the error rate remains below 2%. Compared to the You Only Look Once Version 4 (YOLOv4) model, it exhibits greater performance in detecting surface defects. This advancement may lead to increased productivity and quality control in the manufacturing of aluminum profiles.

References

B. Yang, Z. Liu, G. Duan, and J. Tan, “Mask2defect: A Prior Knowledge-Based Data Augmentation Method for Metal Surface Defect Inspection,” IEEE Transactions on Industrial Informatics, vol. 18, no. 10, pp. 6743-6755, 2022.

J. Sun, P. Wang, Y. K. Luo, and W. Li, “Surface Defects Detection Based on Adaptive Multiscale Image Collection and Convolutional Neural Networks,” IEEE Transactions on Instrumentation and Measurement, vol. 68, no. 12, pp. 4787-4797, 2019.

J. Cao, H. Wu, W. Wang, T. Qasim, and D. Wang, “A Visual Inspection and Classification Method for Camshaft Surface Defects Based on Defect Similarity Measurement,” IEEE Access, vol. 12, pp. 74633-74648, 2024.

Z. Guo and T. Sasayama, “Detection of Surface and Back-Surface Defects on Metal Plate via Rectangular Wave Eddy Current Testing Using Magnetoresistive Sensor,” IEEE Transactions on Magnetics, vol. 59, no. 11, pp. 1-5, 2023.

Q. Li, Z. Luo, H. Chen, and C. Li, “An Overview of Deeply Optimized Convolutional Neural Networks and Research in Surface Defect Classification of Workpieces,” IEEE Access, vol. 10, pp. 26443-26462, 2022.

S. She, X. Zheng, L. Xiong, T. Meng, Z. Zhang, Y. Shao, et al., “Thickness Measurement and Surface-Defect Detection for Metal Plate Using Pulsed Eddy Current Testing and Optimized Res2net Network,” IEEE Transactions on Instrumentation and Measurement, vol. 73, pp. 1-13, 2024.

K. Sun, Q. Wen, and H. Zhou, “Ganster R-CNN: Occluded Object Detection Network Based on Generative Adversarial Nets and Faster R-CNN,” IEEE Access, vol. 10, pp. 105022-105030, 2022.

X. Zheng, S. Zheng, Y. Kong, and J. Chen, “Recent Advances in Surface Defect Inspection of Industrial Products Using Deep Learning Techniques,” The International Journal of Advanced Manufacturing Technology, vol. 113, no. 1-2, pp. 35-58, 2021.

Y. S. Balcıoğlu, B. Sezen, C. C. Çerasi, and S. H. Huang, “Machine Design Automation Model for Metal Production Defect Recognition with Deep Graph Convolutional Neural Network,” Electronics, vol. 12, no. 4, article no. 825, 2023.

B. Li, F. Ren, H. Ni, X. Kang, S. Lv, and Z. Hao, “Classification Method of Surface Defects of Aluminum Profile Based on Transfer Learning,” International Conference on Machine Learning and Intelligent Systems Engineering, pp. 1-5, 2022.

C. Jia and F. Huang, “A Novel Fault Inspection Method of Steel Plate Surface,” 4th International Conference on Artificial Intelligence and Advanced Manufacturing, pp. 66-73, 2022.

J. Y. Choi and J. M. Han, “Deep Learning (Fast R-CNN)-Based Evaluation of Rail Surface Defects,” Applied Sciences, vol. 14, no. 5, article no. 1874, 2024.

L. Yang, X. Huang, Y. Ren, and Q. Han, “Study on Image Enhancement and Automatic Annotation of Steel Plate Surfaced Defect,” Mechanical Science and Technology for Aerospace Engineering, vol. 44, no. 3, pp. 445-452, 2025. (In Chinese)

X. Shi, S. Zhou, Y. Tai, J. Wang, S. Wu, J. Liu, et al., “An Improved Faster R-CNN for Steel Surface Defect Detection,” IEEE 24th International Workshop on Multimedia Signal Processing, pp. 1-5, 2022.

B. Jiang, P. Zhou, X. Sun, and T. Chai, “Intelligent Recognition of Steel Plate Surface Defect Based on Deep Convolutional GAN,” Neural Computing and Applications, vol. 37, no. 14, pp. 8331-8345, 2025.

M. Sun, F. Dong, Z. Huang, and J. Luo, “Adaptive Model Compression for Steel Plate Surface Defect Detection: An Expert Knowledge and Working Condition-Based Approach,” Tsinghua Science and Technology. vol. 29, no. 6, pp. 1851-1871, 2024.

C. Zhang, J. Cui, J. Wu, and X. Zhang, “Attention Mechanism and Texture Contextual Information for Steel Plate Defects Detection,” Journal of Intelligent Manufacturing, vol. 35, no. 5, pp. 2193-2214, 2024.

F. Selamet, S. Cakar, and M. Kotan, “Automatic Detection and Classification of Defective Areas on Metal Parts by Using Adaptive Fusion of Faster R-CNN and Shape from Shading,” IEEE Access, vol. 10, pp. 126030-126038, 2022.

L. Jiang, B. Yuan, Y. Wang, Y. Ma, J. Du, F. Wang, et al., “Ma-Yolo: A Method for Detecting Surface Defects of Aluminum Profiles with Attention Guidance,” IEEE Access, vol. 11, pp. 71269-71286, 2023.

P. Tan, X. Li, Z. Wu, J. Ding, J. Ma, Y. Chen, et al., “Multialgorithm Fusion Image Processing for High Speed Railway Dropper Failure–Defect Detection,” IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 51, no. 7, pp. 4466-4478, 2021.

L. Cui, X. Jiang, M. Xu, W. Li, P. Lv, and B. Zhou, “SDDNet: A Fast and Accurate Network for Surface Defect Detection,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1-13, 2021.

J. Yang, W. Wang, G. Lin, Q. Li, Y. Sun, and Y. Sun, “Infrared Thermal Imaging-Based Crack Detection Using Deep Learning,” IEEE Access, vol. 7, pp. 182060-182077, 2019.

K. Xiang, S. Li, M. Luan, Y. Yang, and H. He, “Aluminum Product Surface Defect Detection Method Based on Improved Faster RCNN,” Chinese Journal of Scientific Instrument, vol. 42, no. 01, pp. 191-198, 2021. (In Chinese)