DOI QR코드

DOI QR Code

전이학습 기반 CNN을 통한 풀림 방지 코팅 볼트 이진 분류에 관한 연구

Binary classification of bolts with anti-loosening coating using transfer learning-based CNN

  • 노은솔 (공주대학교 미래융합공학과) ;
  • 이사랑 (공주대학교 미래융합공학과) ;
  • 홍석무 (공주대학교 미래융합공학과)
  • Noh, Eunsol (Department of Future Convergence Engineering, Kongju National University) ;
  • Yi, Sarang (Department of Future Convergence Engineering, Kongju National University) ;
  • Hong, Seokmoo (Department of Future Convergence Engineering, Kongju National University)
  • 투고 : 2020.10.17
  • 심사 : 2021.02.05
  • 발행 : 2021.02.28

초록

풀림 방지 코팅 볼트는 주로 자동차 안전 관련 부품을 결합하는 데 사용되므로 안전성 유지를 위해 코팅 결함을 사전에 감지해야 한다. 이를 위해 이전 연구 [CNN 및 모델 시각화 기법을 사용한 코팅 볼트 불량 판별]에서는 합성곱 신경망을 사용했다. 이때 합성곱 신경망은 데이터 수가 많을수록 이미지 패턴 및 특성 분석 정확도가 증가하지만 그에 따라 학습시간이 증가한다. 또한 확보 가능한 코팅 볼트 샘플이 한정적이다. 본 연구에서는 이전 연구에 전이학습을 추가적으로 적용해 데이터 개수가 적은 경우에도 코팅 결함에 대해 정확한 분류를 하고자 한다. 전이학습을 적용할 때 학습 데이터 수와 사전 학습 데이터 ImageNet 간의 유사성을 고려해 분류층만 학습했다. 데이터 학습에는 전역 평균 풀링, 선형 서포트 벡터 머신 및 완전 연결 계층과 같은 분류층을 적용했으며, 고려한 모델 중 완전 연결 계층 방법의 분류층이 가장 높은 95% 정확도를 가진다. 추가적으로 마지막 합성곱층과 분류층을 미세 조정하면 정확도는 97%까지 향상된다. 전이학습 및 미세 조정을 이용하면 선별 정확도를 향상시킴은 물론 이전보다 학습 소요시간을 절반으로 줄일 수 있음을 보였다.

Because bolts with anti-loosening coatings are used mainly for joining safety-related components in automobiles, accurate automatic screening of these coatings is essential to detect defects efficiently. The performance of the convolutional neural network (CNN) used in a previous study [Identification of bolt coating defects using CNN and Grad-CAM] increased with increasing number of data for the analysis of image patterns and characteristics. On the other hand, obtaining the necessary amount of data for coated bolts is difficult, making training time-consuming. In this paper, resorting to the same VGG16 model as in a previous study, transfer learning was applied to decrease the training time and achieve the same or better accuracy with fewer data. The classifier was trained, considering the number of training data for this study and its similarity with ImageNet data. In conjunction with the fully connected layer, the highest accuracy was achieved (95%). To enhance the performance further, the last convolution layer and the classifier were fine-tuned, which resulted in a 2% increase in accuracy (97%). This shows that the learning time can be reduced by transfer learning and fine-tuning while maintaining a high screening accuracy.

키워드

참고문헌

  1. E. Noh, S. Yi, M. Kim and S. Hong, "Identification of bolt coating defects using CNN and Grad-CAM", Journal of the Korea Scociety of Mechanical Engineers, 2020, In press
  2. M. K. Ferguson, A. K. Ronay, Y. T. T. Lee and K. H. Law, "Detection and Segmentation of Manufacturing Defects with Convolution Neural Networks and Transfer Learning", Smart Sustainable Manufacturing System, Vol.2, pp.137-164, Sep. 2018. DOI: https://doi.org/10.1520/SSMS20180033
  3. K. Gopalakrishnan, S. K. Khaitan, A. Choudhary and A. Agrawal, "Deep Convolutional Neural Networks with transfer learning for computer vision-based data-driven pavement distress detection", Construction and Building Materials, Vol.157, pp.322-330, Dec. 2017. DOI: https://doi.org/10.1016/j.conbuildmat.2017.09.110
  4. K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition", arXiv:1409.1556, pp.1-14, 2014.
  5. S. H. Lee, "Deep learning based face mask recognition for access control", Journal of the Korea Academia-Industrial cooperation Society, Vol.21, No.8, pp.395-400, Aug. 2020. DOI: http://dx.doi.org/10.5762/KAIS.2020.21.8.395
  6. K. Imoto, T. Nakai, T. Ike, K. Haruki and Y. Sato, "A CNN-based Transfer Learning Method for Defect Classification in Semiconductor Manufacturing", 2018 International Symposium on Semiconductor Manufacturing, Tokyo, Japan, Dec. 2018. DOI: http://dx.doi.org/10.1109/ISSM.2018.8651174
  7. J. Wu, Z. Zhao, C. Sun, R. Yan and X. Chen, "Few-shot transfer learning for intelligent fault diagnosis of machine", Journal of Measurement, Vol.166, No.15, pp.1-13, Dec. 2020. DOI: https://doi.org/10.1016/j.measurement.2020.108202
  8. S. Kim, Y. K. Noh and F. C. Park, "Efficient neural network compression via transfer learning for machine vision inspection", Journal of Neurocomputing, Vol.413, pp.294-304, Nov. 2020. DOI: https://doi.org/10.1016/j.neucom.2020.06.107
  9. S. J. Pan and Q. Yang, "A Survey on Transfer Learning", IEEE Transactions of Knowledge and Data Engineering, Vol.22, No.10, pp.1345-1359, Oct. 2010. DOI: http://dx.doi.org/10.1109/TKDE.2009.191
  10. H. Lee, J. Kim, J. Yu, Y. Jeong and S. Kim, "Multi-class Classification using Transfer Learning based Convolutional Neural Network", Journal of Korean Institute of Intelligent Systems, Vol.28, No.6, pp.531-537, Dec. 2018. DOI: http://dx.doi.org/10.5391/JKIIS.2018.28.6.531
  11. M. Lin, Q. Chen and S. Yan, "Network in network", arXiv:1312.4400, 2014.
  12. Y. Tang, "Deep Learning using Linear Support Vector Machines", arXiv:1306.0239, Jun. 2013.
  13. H. C. Shin, H. R. Roth, M. Gao, L. Lu, Z. Xu, I. Nogues, J. Yao, D. Mollura and R. M. Summers, "Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning", IEEE Transactions on Medical Imaging, Vol.35, No.5, pp.1285-1298, May 2016. DOI: http://dx.doi.org/10.1109/TMI.2016.2528162
  14. N. K. Kim, J. W. Lee, J. I. Kim and S. H. Hong, "Exotic Plants Classification Using the Transfer Learning", The Institute of Electronics and Information Engineers, Vol.41, No.2, pp.477-480, Nov. 2018.
  15. S. Kim, W. Kim, Y. K. Noh and F. C. Park, "Transfer learning for automated optical inspection", 2017 International Joint Conference on Neural Networks, AK, USA, May 2017. DOI: http://dx.doi.org/10.1109/IJCNN.2017.7966162