Classification of Surface Defect on Steel Strip by KNN Classifier

KNN 분류기에 의한 강판 표면 결함의 분류

  • 김철호 (서울산업대학교 대학원 메카트로닉스 공학과) ;
  • 최세호 (포스코 기술연구소 계측연구그룹) ;
  • 김기범 (서울산업대학교 기계설계.자동화공학부) ;
  • 주원종 (서울산업대학교 기계설계.자동화공학부)
  • Published : 2006.08.01

Abstract

This paper proposes a new steel strip surface inspection system. The system acquires bright and dark field images of defects by using a stroboscopic IR LED illuminator and area camera system and the defect images are preprocessed and segmented in real time for feature extraction. 4113 defect samples of hot rolled steel strip are used to develop KNN (k- Nearest Neighbor) classifier which classifies the defects into 8 different types. The developed KNN classifier demonstrates about 85% classifying performance which is considered very plausible result.

Keywords

References

  1. Song, S. J., Kim, H. J., Choi, S. H. and Lee, J. H., 'Classification of Surface Defects on Cold Rolled Strips by Probabilistic Neural Networks,' Journal of the Korean Society for Nondestructive Testing, Vol. 17, No.3, pp. 162-173, 1997
  2. Kim, K. M., Park, G. T., Park, J. J., Lee, J. H., Jung, J. Y. and Lee, J. K., 'Development of Surface Defect Inspection Algorithms for Cold Mill Strip,' Journal of Control Automation and System Engineering, Vol. 3, No. 2, pp. 179-186, 1997.
  3. Lee, S. W., Park, W. J., Joo, J. I., Jung, J. Y., Lee, J. H. and Park, C. H., 'Development of Surface Inspection ?System for # I PCM,' Technical Research Lab. POSCO, 2000
  4. Gonzalez, R. C. and Woods, R. E., 'Digital Image Processing,' Addison-Wesley, 1992
  5. Randy, C., 'A simplified approach to image processing,' Prentice Hall, 1996
  6. Gose, E., Honsonbaugh, R. and Jost, S., 'Pattern Recognition and Image Analysis,' Prentice Hall, 1996
  7. Umbaugh, S. E., 'Digital Image Analysis and Processing,' CRC Press, 2005
  8. Choi, Y. I., Park, K. H. and Kee, C. D., 'Rejection Scheme of Nearest Neighbor Classifier for Diagnosis of Rotating Machine Fault,' Journal of the Korean Society of Precision Engineering, Vol. 19, No. 3, pp. 52-58, 2000