Resources Recycling (자원리싸이클링)

The Korean Institute of Resources Recycling (한국자원리싸이클링학회)
권호 표지
DOI QR코드

DOI QR Code

Estimated Soft Information based Most Probable Classification Scheme for Sorting Metal Scraps with Laser-induced Breakdown Spectroscopy

레이저유도 플라즈마 분광법을 이용한 폐금속 분류를 위한 추정 연성정보 기반의 최빈 분류 기술

  • Kim, Eden (School of Mechanical Engineering, Gwangju Institute of Science and Technology) ;
  • Jang, Hyemin (School of Mechanical Engineering, Gwangju Institute of Science and Technology) ;
  • Shin, Sungho (School of Mechanical Engineering, Gwangju Institute of Science and Technology) ;
  • Jeong, Sungho (School of Mechanical Engineering, Gwangju Institute of Science and Technology) ;
  • Hwang, Euiseok (School of Mechanical Engineering, Gwangju Institute of Science and Technology)
  • 김에덴 (광주과학기술원 기계공학부) ;
  • 장혜민 (광주과학기술원 기계공학부) ;
  • 신성호 (광주과학기술원 기계공학부) ;
  • 정성호 (광주과학기술원 기계공학부) ;
  • 황의석 (광주과학기술원 기계공학부)
  • Received : 2018.01.11
  • Accepted : 2018.02.02
  • Published : 2018.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, a novel soft information based most probable classification scheme is proposed for sorting recyclable metal alloys with laser induced breakdown spectroscopy (LIBS). Regression analysis with LIBS captured spectrums for estimating concentrations of common elements can be efficient for classifying unknown arbitrary metal alloys, even when that particular alloy is not included for training. Therefore, partial least square regression (PLSR) is employed in the proposed scheme, where spectrums of the certified reference materials (CRMs) are used for training. With the PLSR model, the concentrations of the test spectrum are estimated independently and are compared to those of CRMs for finding out the most probable class. Then, joint soft information can be obtained by assuming multi-variate normal (MVN) distribution, which enables to account the probability measure or a prior information and improves classification performance. For evaluating the proposed schemes, MVN soft information is evaluated based on PLSR of LIBS captured spectrums of 9 metal CRMs, and tested for classifying unknown metal alloys. Furthermore, the likelihood is evaluated with the radar chart to effectively visualize and search the most probable class among the candidates. By the leave-one-out cross validation tests, the proposed scheme is not only showing improved classification accuracies but also helpful for adaptive post-processing to correct the mis-classifications.

본 연구에서는 레이저유도 플라즈마 분광법(Laser induced breakdown spectroscopy, LIBS) 기반의 금속 종류별 스펙트럼 데이터를 이용하여 연성정보(soft information)를 추정하고 최빈 클래스로 분류하는(most probable classification) 금속 분류 방법을 제안한다. 폐금속 자원과 같이 사전 정보가 없는 금속을 분류하는 경우 몇 가지 핵심 구성성분에 대한 정량 분석을 통해서 클래스를 추정하는 방법이 효율적이다. 이에 따라 부분 집합 기반의 부분최소제곱회귀법(Partial Least Square Regression, PLSR)을 이용하여 LIBS 검출 스펙트럼으로부터 각 성분의 농도를 독립적으로 신뢰성 있게 추정하고, 인증 표준물질(CRM) 등 알려진 모집합의 농도정보에 기반하여 최고 확률을 갖도록 분류하는 기술을 제안한다. 샘플 스펙트럼들의 다변량 분석을 통해서 여러 성분의 추정 농도를 다변량 정규 분포를 갖는 것으로 가정하고 통합(Joint) 추정 연성정보를 구할 수 있으며, 이를 활용한 최빈 확률 검출이나 추가적인 사전 정보의 결합 등을 통해서 분류 성능을 향상시킬 수 있다. 제안된 기술의 평가를 위해서 9가지 종류의 CRM 금속시료의 LIBS 스펙트럼 데이터를 사용하며, 부분 집합 기반의 PLSR 농도 추정 기술을 기반으로 단변량 혹은 다변량 정규 분포 연성 정보추정을 통해 미지 금속의 검출과 연성 정보의 검출 등을 테스트 하였다. 또한 방사형 차트(Radar chart)를 이용하여 추정된 농도와 획득한 연성정보를 효과적으로 시각화함으로써 기존 라이브러리에 포함된 부분 집합의 금속과 비교하여 해당 금속과의 유사성을 그래프를 통해 추정할 수 있다.

Keywords

References

  1. Gurell, J., et al., 2012 : Laser induced breakdown spectroscopy for fast elemental analysis and sorting of metallic scrap pieces using certified reference materials, Spectrochim. Acta, Part B., 74-75, pp.46-50. https://doi.org/10.1016/j.sab.2012.06.013
  2. Choi, H. K. et al., 2007 : Technology development of separation for waste plastics by the near-infrared spectra method, KIC News, 10(3), pp.22-41.
  3. Koyanak, S. and Kobayashi, K., 2010 : Automatic sorting of lightweight metal scrap by sensing apparent density and three-dimensional shape, Resour Conserv Recycl., 54, pp.571-578. https://doi.org/10.1016/j.resconrec.2009.10.014
  4. Jung, I. S. et al., 2015 : Overview and recent development of recycling small waste electrical and electronic equipment (WEEE), J. of Korean Inst. of Resources Recycling, 24(4), pp.38-49. https://doi.org/10.7844/kirr.2015.24.4.38
  5. Cremers, D. A. and Radziemski, L. J., 2006 : Handbook of laser-induced breakdown spectroscopy, pp.23-53, 2nd Edition, John Wiley & Sons Ltd, England.
  6. Salle, Beatrice, et al., 2014 : Laser-induced breakdown spectroscopy for Mars surface analysis: capabilities at stand-off distances and detection of chlorine and sulfur elements, Spectrochimica Acta Part B: Atomic Spectroscopy, 59(9), pp.1413-1422. https://doi.org/10.1016/j.sab.2004.06.006
  7. Kumar, Akshaya, et al., 2004 : Characterization of malignant tissue cells by laser-induced breakdown spectroscopy, Applied optics, 43(28), pp.5399-5403. https://doi.org/10.1364/AO.43.005399
  8. Gaudiuso, Rosalba, et al., 2010 : Laser induced breakdown spectroscopy for elemental analysis in environmental, cultural heritage and space applications: a review of methods and results, Sensors, 10(8), pp.7434-7468. https://doi.org/10.3390/s100807434
  9. Anabitarte, F., A. Cobo, and Jose Miguel Lopez-Higuera., 2012 : Laser-induced breakdown spectroscopy: fundamentals, applications, and challenges, ISRN Spectroscopy 2012.
  10. In, Jung-Hwan, et al., 2013 : Improvement of selenium analysis during laser-induced breakdown spectroscopy measurement of Culn 1-x Ga x Se2 solar cell films by self-absorption corrected normalization, Journal of Analytical Atomic Spectrometry, 28(8), pp.1327-1337. https://doi.org/10.1039/c3ja50079a
  11. Syvilay, D., et al., Trace element quantification of lead based roof sheets of historical monuments by laser induced breakdown spectroscopy, Spectrochimica Acta Part B: Atomic Spectroscopy, 103, pp.34-42.
  12. Boucher, Thomas F., et al., 2015 : A study of machine learning regression methods for major elemental analysis of rocks using laser-induced breakdown spectroscopy, Spectrochimica Acta Part B: Atomic Spectroscopy, 107, pp.1-10. https://doi.org/10.1016/j.sab.2015.02.003
  13. Clegg, Samuel M., et al., 2009 : Multivariate analysis of remote laser-induced breakdown spectroscopy spectra using partial least squres, principal component analysis, and related techniques, Spectrochimica Acta Part B: Atomic Spectroscopy, 64(1), pp.79-88. https://doi.org/10.1016/j.sab.2008.10.045
  14. Brammer standard reference materials, http://www.brammerstandard.com/
  15. Shin, S. H., et al., 2016 : Multivariate analysis for classification of metals by laser induced breakdown spectroscopy, Spring Annual Proc. Korean Soc. of Laser Processing.
  16. Jang, H. M., et al., 2018 : Most informative Spectral Line Processing Scheme for Laser-induced Breakdown Spectroscopy based Classification of Same-base Alloys, Submitted to Spectrochimica Acta Part B: Atomic Spectroscopy