Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Design of Automatic Classification System of Black Plastics Based on Support Vector Machine Using Raman Spectroscopy

라만분광법을 이용한 SVM 기반 흑색 플라스틱 자동 분류 시스템의 설계

  • Bae, Jong-Soo (Department of Electrical Engineering, The University of Suwon) ;
  • Oh, Sung-Kwun (Department of Electrical Engineering, The University of Suwon) ;
  • Kim, Hyun-Ki (Department of Electrical Engineering, The University of Suwon)
  • Received : 2016.05.20
  • Accepted : 2016.10.13
  • Published : 2016.10.25
Download PDF
⟨ Previous Next ⟩

Abstract

Lots of plastics are widely used in a variety of industrial field. And the amount of plastic waste is massively produced. In the study of waste recycling, it is emerged as an important issue to prevent the waste of potentially useful resource materials as well as to reduce ecological damage. So, the recycling of plastic waste has been currently paid attention to from the view point of reuse. Existing automatic sorting system consist of near infrared ray (NIR) sensors to classify the types of plastics. But the classification of black plastics still remains a challenge. Black plastics which contains carbon black are not almost classified by NIR because of the characteristic of the light absorption of black plastics. This study is focused on handling how to identify black plastics instead of NIR. Raman spectroscopy is used to get qualitative as well as quantitative analysis of black plastics. In order to improve the performance of identification, Support Vector Machine(SVM) classifier and Principal Component Analysis(PCA) are exploited to more preferably classify some kinds of the black plastics, and to analyze the characteristic of each data.

수많은 플라스틱이 산업분야에 다양하게 사용되어지고 있다. 또한 많은 양의 플라스틱 폐기물들이 발생하고 있다. 재활용에 대한 연구는 환경오염 뿐만아니라 한정된 유용한 자원이 버려지는 것을 방지하기 위해 중요한 이슈로 부각되고 있다. 이렇기 때문에 폐플라스틱의 재활용은 재사용 관점에서 주목받고 있는 실정이다. 현재 재활용 센터에서는 플라스틱의 재질을 분류하기 위해 NIR 센서를 이용한 플라스틱 자동 분류 시스템을 구축 및 운용하고 있다. 하지만 흑색 플라스틱은 여전히 분류가 되지 않는 도전적인 목표로 남아있다. 카본 블랙이 포함된 흑색 플라스틱의 경우 검정색의 특성상 NIR 장비에서 나오는 빛을 흡수하기 때문에 분류에 어려움이 있다. 본 연구는 NIR 장비 대신 흑색 플라스틱을 분류하는 방법에 대한 연구이다. 흑색 플라스틱의 정성적, 정량적 분석을 위해 Raman 분광법을 사용하였다. 또한 분류기의 인식률을 높이기 위해 데이터를 특성을 분석하고 흑색 플라스틱을 좀 더 확실하게 분류하기 위해 Support Vector Machine(SVM), 주성분 분석법(PCA) 같은 알고리즘을 이용하였다.

Keywords

References

  1. B.D. Hardesty, W.Chris, "Eight million tonnes of plastic are going into the ocean each year". The Conversation, Retrieved 21 February 2015.
  2. R.K. Khanna, "Raman-spectroscopy of oligomeric SiO species isolated in solid methane". Journal of Chemical Physics 74 (4): 2108. Bibcode:1 981JChPh..74.2108K. doi:10.1063/1.441393, 1981.
  3. H.S. Chung, "The Optical Analysis of Graphene-Focusing on Raman Spectrum", Physics & Advanced Scientific Technology, 18.7-8, p20-25, 2009.
  4. M.A.De Baez, P.J.Hendra, M.Judkins, "The Raman spectra of oriented isotactic polypropylene", Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 51.12: 2117-2124, 1995. https://doi.org/10.1016/0584-8539(95)01512-1
  5. E.Andreassen, "Infrared and Raman spectroscopy of polypropylene. In: Polypropylene", Springer Netherlands, p. 320-328, 1999.
  6. J. R. Anema, A. G. Brolo, A. Felten, C.Bittencourt, "Surfaceenhanced Raman scattering from polystyrene on gold clusters", Journal of Raman Spectroscopy, 41.7: 745-751, 2010.
  7. D.H.Zhang, J.G.Qin, J.S.Shen, Y.Wang, W.J.Liu, "Study on the concentration dependence of orientation of polystyrene on silver by the sers technique", 18(2), 177-180, 2000.
  8. M.Mazilu, C.D.L.Anna, A.Riches, C. S.Herrington, K.Dholakia, "Optimal algorithm for fluorescence suppression of modulated Raman spectroscopy", Optics express, 18.11: 11382-11395, 2010. https://doi.org/10.1364/OE.18.011382
  9. GnoSys Global Ltd, "PET Analysis", GnoSys Global Ltd, Guildford, Surrey, TSAN11-Application notes.
  10. Peter J. B. Hancock, A. Mike Burton, and Vicki Bruce. "Face processing: Human perception and principal components analysis," Memory and Cognition, Volume: 24, Issue: 1, pp.26-40, 1996. https://doi.org/10.3758/BF03197270
  11. S-H.Yoo, S-K.Oh, P. Witold, "Design of face recognition algorithm using PCA-LDA combined for hybrid data preprocessing and polynomial-based RBF neural networks: Design and its application", Expert Systems with Applications, 40.5: 1451-1466, 2013. https://doi.org/10.1016/j.eswa.2012.08.046
  12. V. Vapnik, "The Nature of Statistical Learning Theory, "Springer Verlag, heidelberg, DE, 1995.
  13. T. Joachims, "SVM Light, Support Vector Machine 2008.
  14. H-S. Han, U-P. Chong, "Electroencephalogram-based Driver Drowsiness Detection System Using AR Coefficients and SVM", Korea Intelligent information System Society journals, 22(6), 768-773, 2012.

Cited by

  1. Design of Pattern Classifier for Electrical and Electronic Waste Plastic Devices Using LIBS Spectrometer vol.26, pp.6, 2016, https://doi.org/10.5391/JKIIS.2016.26.6.477