Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지

Surface Flaw Detection of Cold-Rolled Steel Strips using Intensity Gradient

광강도차를 이용한 냉연강판 표면결함 검출

  • 공선곤 (숭실대학교 전기공학과)
  • Published : 2000.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents a method of detecting surface flaw of cold-rolled steel plate using image processing technique and a neural network classifier. The amount of steel plate surface image data is reduced by the wavelet transform. Features are extracted from the co-occurence matrix of the partial image corresponding to the low-frequency region, and a MLP neural network classifies into predetermined surface flaw categories. Simulations show the neural network classifier outperforms conventional vector quantization method.

이논문에서는 영상처리 기법과 신경회로망 분류기를 이용하여 냉연강판의 표면에 나타나는 결함을 검출하는 기법을 제안한다. 냉연강판 표면의 입력영상을 웨이블렛 변환하여 영상신호 데이터량을 감소시키고 영상신호의 저주파수 영역에 해당하는 부분영상으로부터 구한 co-occurrence 행렬을 이용하여 주된 특징들을 추출한 후 신경회로망 분류기를 이용하여 표면결함을 분류하는 과정을 밟는다 현장에서 직접입수한 실제 냉연강판 표면결함 영상에 대하여 결함의 검출 및 분류기법을 제시하고 실험을 통해 기존의 백터양자화 기법과 비교하여 우수한 성능을 보임을 입증하였다.

Keywords

References

  1. Int. Conf. on Automation, Robotics and Computer Vision Automated visual inspection of painted metallic surfaces A. Ercil
  2. On-line automatic defect detection and surface roughness measurement of steel strip F. Treiber
  3. IEEE Trans. on Pattern Analysis and Machine Intelligence v.10 no.1 An automatic wafer inspection system using pipelined image processing technique H. Yoda.;Y. Chuchi;Y. Taniguchi;M. Ejiri
  4. Computer Vision and Image Understanding v.61 no.2 A survey of automated visual inspection T.S. Newman
  5. Proc. Int. Conf. on Image Processing Visual system for on-line surface inspection in aluminum casting process C. Fernandez;C. Paltero;P. Campoy;R. Arail
  6. CVGIP : Image Understanding v.59 no.3 Image analysis and computer vision : 1993 A. Rosenfeld
  7. Pattern Recognition v.25 no.3 Unsupervised textual classification of image using the texture spectrum D.C. Heand;L. Wang
  8. Pattern Recognition v.25 no.9 An efficient approach to estimate fractal dimensions of textural images N. Sarkar;B.B. Chaudhuri
  9. Pattern Recognition v.25 no.7 Performing the theory of invariant moment for pattern recognition Y. Li
  10. IAPR Int. Conf. on Pattern Recognition v.3 Development environment for designing and testing inspection systems D. Brzakovic;N. Vujovic
  11. IEEE Trans. on Neural Networks v.6 no.5 A skeleton and neural-network based approaches for identifying cosmetic surface flaws C. Wang;D.J. Cannon;S.R.T. Kumara;G. Lu
  12. IEEE Int. Joint Conf. on Neural Networks Syntactic pattern recognition by quadratic neural nets, a case study : Rail flaw classification P. McKenzie;M. Alder