Journal of information and communication convergence engineering

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Toward Practical Augmentation of Raman Spectra for Deep Learning Classification of Contamination in HDD

  • Received : 2023.03.31
  • Accepted : 2023.06.08
  • Published : 2023.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Deep learning techniques provide powerful solutions to several pattern-recognition problems, including Raman spectral classification. However, these networks require large amounts of labeled data to perform well. Labeled data, which are typically obtained in a laboratory, can potentially be alleviated by data augmentation. This study investigated various data augmentation techniques and applied multiple deep learning methods to Raman spectral classification. Raman spectra yield fingerprint-like information about chemical compositions, but are prone to noise when the particles of the material are small. Five augmentation models were investigated to build robust deep learning classifiers: weighted sums of spectral signals, imitated chemical backgrounds, extended multiplicative signal augmentation, and generated Gaussian and Poisson-distributed noise. We compared the performance of nine state-of-the-art convolutional neural networks with all the augmentation techniques. The LeNet5 models with background noise augmentation yielded the highest accuracy when tested on real-world Raman spectral classification at 88.33% accuracy. A class activation map of the model was generated to provide a qualitative observation of the results.

Keywords

References

  1. G. Guo, C. Bi, and A. A. Mamun, Hard Disk Drive: Mechatronics and Control, FL, Boca Raton: CRC, 2006.
  2. R. Nagarajan, "Survey of cleaning and cleanliness measurement in disk drive manufacture," Precision Cleaning, pp. 13-21, Feb. 1997.
  3. A. Rosenkranz, L. Freeman, B. Suen, Y. Fainman, and F. E. Talke, "Tip-enhanced Raman spectroscopy studies on amorphous carbon films and carbon overcoats in commercial hard disk drives," Tribology Letters, vol. 66, no. 2, pp. 1-6, Mar. 2018. DOI: 10.1007/s11249-018-1005-2.
  4. M. Kansiz, C. Prater, E. Dillon, M. Lo, J. Anderson, C. Marcott, A. Demissie, Y. Chen, and G. Kunkel, "Optical photothermal infrared microspectroscopy with simultaneous Raman - A new non-contact failure analysis technique for identification of <10 ㎛ organic contamination in the hard drive and other electronics industries," Microscopy Today, vol. 28, no. 3, pp. 26-36, May 2020. DOI: 10.1017/S1551929520000917.
  5. X. Fan, W. Ming, H. Zeng, Z. Zhang, and H. Lu, "Deep learning-based component identification for the Raman spectra of mixtures," Analyst, vol. 144, no. 5, pp. 1789-1798, Jan. 2019. DOI: 10.1039/C8AN02212G.
  6. X. Zhang, T. Lin, J. Xu, X. Luo, and Y. Ying, "DeepSpectra: An end-to-end deep learning approach for quantitative spectral analysis," Analytica Chimica Acta, vol. 1058, pp. 48-57, Jun. 2019. DOI: 10.1016/j.aca.2019.01.002.
  7. W. Zhang, W. Feng, Z. Cai, H. Wang, Q. Yan, and Q. Wang, "A deep one-dimensional convolutional neural network for microplastics classification using Raman spectroscopy," Vibrational Spectroscopy, vol. 124, p. 103487, Jan. 2023. DOI: 10.1016/j.vibspec.2022.103487.
  8. X. Qiu, X. Wu, X. Fang, Q. Fu, P. Wang, X. Wang, S. Li, and Y. Li, "Raman spectroscopy combined with deep learning for rapid detection of melanoma at the single cell level," Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, vol. 286, p. 122029, Feb. 2023. DOI: 10.1016/j.saa.2022.122029.
  9. S. Gulyanon, S. Deepaisarn, C. Srisumarnk, N. Chiewnawintawat, A. Angkoonsawaengsuk, S. Laitrakun, P. Opaprakasit, P. Rakpongsiri, T. Meechamnan, and D. Sompongse, "A comparative study of noise augmentation and deep learning methods on Raman spectral classification of contamination in hard disk drive," in 2022 17th International Joint Symposium on Artificial Intelligence and Natural Language Processing (iSAI-NLP), Chiang Mai, Thailand, pp. 1-6, 2022. DOI: 10.1109/iSAI-NLP56921.2022.9960277.
  10. A. Geron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, 3rd ed., Sebastopol, CA: O'Reilly Media, 2022.
  11. Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998. DOI: 10.1109/5.726791.
  12. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Communications of the ACM, vol. 60, no. 6, pp. 84-90, Jun. 2017. DOI: 10.1145/3065386.
  13. K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv 2014. [Online] Available: https://arxiv.org/abs/1409.1556.
  14. C. Szegedy, W. Liu, Y. Jia, P. Semanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, "Going deeper with convolutions," arXiv 2014. [Online] Available: https://arxiv.org/abs/1409.4842.
  15. K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas: NV, pp. 770-778, 2016. DOI: 10.1109/CVPR.2016.90.
  16. F. N. Iandola, S. Han, M. W. Moskewicz, K. Ashraf, W. J. Dally, and K. Keutzer, "SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size," arXiv 2016. [Online] Available: https://arxiv.org/abs/1602.07360.
  17. F. Chollet, "Xception: Deep learning with depthwise separable convolutions," arXiv 2016. [Online] Available: https://arxiv.org/abs/1610.02357.
  18. G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, "Densely connected convolutional networks," in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu: HI, pp. 4700-4708, 2017. DOI: 10.1109/cvpr.2017.243.
  19. A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, M. Andreetto, and H. Adam, "MobileNets: Efficient convolutional neural networks for mobile vision applications," arXiv 2017. [Online] Available: https://arxiv.org/abs/1704.04861.
  20. C.-S. Ho, N. Jean, C. A. Hogan, L. Blackmon, S. S. Jeffrey, M. Holodniy, N. Banaei, A. A. E. Saleh, S. Ermon, and J. Dionne, "Rapid identification of pathogenic bacteria using Raman spectroscopy and deep learning," Nature Communications, vol. 10, p. 4927, Oct. 2019. DOI: 10.1038/s41467-019-12898-9.
  21. C. Chen, W. Wu, C. Chen, F. Chen, X. Dong, M. Ma, Z. Yan, X. Lv, Y. Ma, and M. Zhu, "Rapid diagnosis of lung cancer and glioma based on serum Raman spectroscopy combined with deep learning," Journal of Raman Spectroscopy, vol. 52, no. 11, pp. 1798-1809, Aug. 2021. DOI: 10.1002/jrs.6224.
  22. X. Zhang, X. Song, W. Li, C. Chen, M. Wusiman, L. Zhang, J. Zhang, J. Lu, C. Lu, and X. Lv, "Rapid diagnosis of membranous nephropathy based on serum and urine Raman spectroscopy combined with deep learning methods," Scientific Reports, vol. 13, p. 3418, Feb. 2023. DOI: 10.1038/s41598-022-22204-1.
  23. X. Chang, M. Yu, R. Liu, R. Jing, J. Ding, J. Xia, Z. Zhu, X. Li, Q. Yao, L. Zhu, and T. Zhang, "Deep learning methods for oral cancer detection using Raman spectroscopy," Vibrational Spectroscopy, vol. 126, p. 103522, May 2023. DOI: 10.1016/j.vibspec.2023.103522.
  24. J. Houston, F. G. Glavin, and M. G. Madden, "Robust classification of high-dimensional spectroscopy data using deep learning and data synthesis," Journal of Chemical Information and Modeling, vol. 60, no. 4, pp. 1936-1954, Mar. 2020. DOI: 10.1021/acs.jcim.9b01037.
  25. M. Wu, S. Wang, S. Pan, A. C. Terentis, J. Strasswimmer, and X. Zhu, "Deep learning data augmentation for Raman spectroscopy cancer tissue classification," Scientific Reports, vol. 11, p. 23842, Dec. 2021. DOI: 10.1038/s41598-021-02687-0.
  26. J. Zhao, H. Lui, D. I. McLean, and H. Zeng, "Automated autofluorescence background subtraction algorithm for biomedical Raman spectroscopy," Applied Spectroscopy, vol. 61, no. 11, pp. 1225-1232, Nov. 2007. DOI: 10.1366/000370207782597003.
  27. U. Blazhko, V. Shapaval, V. Kovalev, and A. Kohler, "Comparison of augmentation and pre-processing for deep learning and chemometric classification of infrared spectra," Chemometrics and Intelligent Laboratory Systems, vol. 215, p. 104367, Aug. 2021. DOI: 10.1016/j.chemolab.2021.104367.
  28. N. K. Afseth and A. Kohler, "Extended multiplicative signal correction in vibrational spectroscopy, a tutorial," Chemometrics and Intelligent Laboratory Systems, vol. 117, pp. 92-99, Aug. 2012. DOI: 10.1016/j.chemolab.2012.03.004.
  29. J. Salmon, Z. Harmany, C.-A. Deledalle, and R. Willett, "Poisson noise reduction with non-local PCA," Journal of Mathematical Imaging and Vision, vol. 48, no. 2, pp. 279-294, Feb. 2014. DOI: 10.1007/s10851-013-0435-6.
  30. H. Paik, N. Sastry, and I. SantiPrabha, "Effectiveness of noise jamming with white gaussian noise and phase noise in amplitude comparison monopulse radar receivers," in 2014 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT), Bangalore, India, pp. 1-5, 2014. DOI: 10.1109/conecct.2014.6740286.
  31. B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, "Learning deep features for discriminative localization," in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas: NV, pp. 2921-2929, 2016. DOI: 10.1109/cvpr.2016.319.
  32. R. L. Draelos and L. Carin, "Use HiResCAM instead of Grad-CAM for faithful explanations of convolutional neural networks," arXiv 2020. [Online] Available: https://arxiv.org/abs/2011.08891.