Defect Detection Approach in Manufacturing Environments using Customizable Convolutional Neural Networks with Multi-Scale Attention Mechanisms
Keywords:
Machine Learning, NAS-ASDet, Conventional Image Processing, AENet, Segmentation, Multi-Scale, Data-Driven, Experimental ResultsAbstract
Three primary issues have hindered the ability of quality control systems to identify superficial faults in industrial manufacturing processes: fuzzy edges, fault characteristics with several scales, and geometric elements that are hard to identify. This paper suggests a surface defect detection technique based on combining pixel-level semantic segmentation with multi-scale data. Industrial defect identification requires real-time and high accuracy in complex, dynamic environments. Conventional image processing & machine learning focused on handcrafted features struggle to meet these needs. The answer to this issue proposed by this research is AENet, a revolutionary immediate form of defect identification network based on an encoder-decoder architecture. In addition to having excellent detection accuracy and efficiency, AENet also shows strong convergence and generalization. The method known as Neural Architecture Search (NAS) allows networks that are driven by data to autonomously generate and adapt. Here, we provide a novel method for surface defect identification using network adaptive design, called NAS-ASDet. This method can be applied to industrial settings to build low-data-weight, high-performance defect detection networks. Based on four datasets, the experimental results demonstrate that the suggested approach achieves a smaller model size and better performance compared to competing methods like manual and NAS-based techniques
Downloads
References
C. Xie, C. Xia, M. Ma, Z. Zhao, X. Chen, and J. Li, “Pyramid grafting network for one-stage high resolution saliency detection,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 11717–11726, 2022.
Y. He, K. Song, Q. Meng, and Y. Yan, “An end-to-end steel surface defect detection approach via fusing multiple hierarchical features,” IEEE Transactions on Instrumentation and Measurement, vol. 69, no. 4, pp. 1493–1504, 2019.
B. Wei, K. Hao, L. Gao, and X.-s. Tang, “Detecting textile microdefects: A novel and efficient method based on visual gain mechanism,” Information Sciences, vol. 541, pp. 60–74, 2020.
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.
B. Su, H. Chen, K. Liu, and W. Liu, “Rcag-net: Residual channelwise attention gate network for hot spot defect detection of photovoltaic farms,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1– 14, 2021.
Y. Tu, Z. Ling, S. Guo, and H. Wen, “An accurate and real-time surface defects detection method for sawn lumber,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–11, 2020.
Y. Huang, C. Qiu, X. Wang, S. Wang, and K. Yuan, “A compact convolutional neural network for surface defect inspection,” Sensors, vol. 20, no. 7, p. 1974, 2020.
D. Zhang, K. Song, J. Xu, Y. He, M. Niu, and Y. Yan, “Mcnet: Multiple context information segmentation network of no-service rail surface defects,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–9, 2021.
Y. Li and J. Li, “An end-to-end defect detection method for mobile phone light guide plate via multitask learning,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–13, 2021.
Gu, Z., Chen, H., Xu, Z., Lan, J., Meng, C., Wang, W.: DiffusionInst: Diffusion Model for Instance Segmentation (2022)
Yuan, X.-c., Wu, L.-s., Peng, Q.: An improved otsu method using the weighted object variance for defect detection. Applied Surface Science 349, 472–484 (2015).
Boukdir, A., Benaddy, M., Meslouhi, O.E., Kardouchi, M., Akhloufi, M.: Character-level arabic text generation from sign language video using encoderdecoder model. Displays 76, 102340 (2022).
Lin, Y.: Wafer pattern counting, detection and classification based on encoderdecoder cnn structure. In: 2022 Intermountain Engineering, Technology and Computing (IETC), pp. 1–5 (2022).
Wang, S., Wang, H., She, S., Zhang, Y., Qiu, Q., Xiao, Z.: Swin-t-nfc crfs: An encoder–decoder neural model for high-precision uav positioning via point cloud super resolution and image semantic segmentation. Comput. Commun. 197(C), 52–60 (2023).
Jie, F., Nie, Q., Li, M., Yin, M., Jin, T.: Atrous spatial pyramid convolution for object detection with encoder-decoder. Neurocomputing 464, 107–118 (2021).
Wang, C., Wang, H.: Cascaded feature fusion with multi-level self-attention mechanism for object detection. Pattern Recognition 138, 109377 (2023).
Sun, J., Wang, P., Luo, Y. K., & Li, W. (2019). Surface defects detection based on adaptive multiscale image collection and convolutional neural networks. IEEE Transactions on Instrumentation and Measurement, 68(12), 4787-4797.
Hou, X., Liu, M., Zhang, S., Wei, P., & Chen, B. (2023). CANet: Contextual Information and Spatial Attention Based Network for Detecting Small Defects in Manufacturing Industry. Pattern Recognition, 140, 109558.
Wan, Y., Yi, L., Jiang, B., Chen, J., Jiang, Y., & Xie, X. (2024). AENet: attention enhancement network for industrial defect detection in complex and sensitive scenarios. The Journal of Supercomputing, 1-24.
Wang, P., Yang, Y., & Moghaddam, N. S. (2022). Process modeling in laser powder bed fusion towards defect detection and quality control via machine learning: The state-of-the-art and research challenges. Journal of Manufacturing Processes, 73, 961-984.
Wang, Y., Hu, S., Wang, G., Chen, C., & Pan, Z. (2020). Multi-scale dilated convolution of convolutional neural network for crowd counting. Multimedia Tools and Applications, 79, 1057-1073.
Huang, C., Wu, X., Zhang, X., Zhang, C., Zhao, J., Yin, D., & Chawla, N. V. (2019, July). Online purchase prediction via multi-scale modeling of behavior dynamics. In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining (pp. 2613-2622).
Qi, X., Li, K., Liu, P., Zhou, X., & Sun, M. (2020). Deep attention and multi-scale networks for accurate remote sensing image segmentation. IEEE Access, 8, 146627-146639.
Xiang, L., Wang, P., Yang, X., Hu, A., & Su, H. (2021). Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism. Measurement, 175, 109094.
Ravindranath, M., Candan, K. S., & Sapino, M. L. (2020, October). M2NN: Rare event inference through multi-variate multi-scale attention. In 2020 IEEE International Conference on Smart Data Services (SMDS) (pp. 53-62). IEEE.
Srivastava, A., Chanda, S., & Pal, U. (2021, November). Exploiting multi-scale fusion, spatial attention and patch interaction techniques for text-independent writer identification. In Asian Conference on Pattern Recognition (pp. 203-217). Cham: Springer International Publishing.
Kong, W., Liu, S., Xu, M., Yasir, M., Wang, D., & Liu, W. (2023). Lightweight algorithm for multi-scale ship detection based on high-resolution SAR images. International Journal of Remote Sensing, 44(4), 1390-1415.
Xu, C., Zhang, S., Zhao, B., Liu, C., Sui, H., Yang, W., & Mei, L. (2022). SAR image water extraction using the attention U-net and multi-scale level set method: flood monitoring in South China in 2020 as a test case. Geo-Spatial Information Science, 25(2), 155-168.
Jiao, J., Zhao, M., Lin, J., & Liang, K. (2020). A comprehensive review on convolutional neural network in machine fault diagnosis. Neurocomputing, 417, 36-63.
Guo, R.; Wang, T.; Li, Y.T. Health Status Recognition of Gear Pump Based on Bayesian-LSTM. J/OL. J. Mech. Eng. 2023, 59, 1–9.
Rana, A.S.; Thomas, M.S.; Senroy, N. Reliability evaluation of WAMS using Markov-based graph theory approach. Iet Gener. Transm. Distrib. 2017, 11, 2930–2937.
Zhao, R.; Yan, R.; Chen, Z. Deep learning and its applications to machine health monitoring. Mech. Syst. Signal Process. 2019, 115, 213–237.
Chen, R.X.; Zhang, Y.; Hu, X.L. Health state recognition of harmonic reducer based on depth feature learning of voltage signal. Chin. J. Sci. Instrum. 2021, 42, 234–241.
Chen, Z.Q.; Chen, X.D.; José, V.D.O. Application of deep learning in equipment prognostics and health management. Chin. J. Sci. Instrum. 2019, 40, 206–226.
Lei, Y.G.; Jia, F.; Zhou, X. Health monitoring method for big data of mechanical equipment based on deep learning theory. J. Mech. Eng. 2015, 51, 49–56.
Chen, H.P.; Hu, N.Q.; Cheng, Z. A deep convolutional neural network based fusion method of two-direction vibration signal data for health state identification of planetary gearboxes. Measurement 2019, 146, 268–278.
Zhang, Y.; Feng, Q.Q.; Huang, Q.Q. Rolling bearing state recognition under variable condition using part-based representation of no negativity constrained auto encoder networks. Chin. J. Sci. Instrum. 2020, 41, 77–85.
Chen, R.X.; Chen, S.Y.; Hu, X.L. Life status identification of rolling bearings under different working conditions by multi-source integrated GFK. J. Vib. Eng. 2020, 33, 614–621..
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
