CFP last date
20 June 2024
Reseach Article

Laws based Quality Inspection of Steel Products using Scanning Electron Microscopy (SEM) Images

Published on May 2012 by B. Sathya Bama, S. Ramabai, R. Santhanadevi
National Conference on Advances in Computer Science and Applications (NCACSA 2012)
Foundation of Computer Science USA
NCACSA - Number 5
May 2012
Authors: B. Sathya Bama, S. Ramabai, R. Santhanadevi
dd2b3367-9b27-459f-a602-6fc8b5dad68d

B. Sathya Bama, S. Ramabai, R. Santhanadevi . Laws based Quality Inspection of Steel Products using Scanning Electron Microscopy (SEM) Images. National Conference on Advances in Computer Science and Applications (NCACSA 2012). NCACSA, 5 (May 2012), 25-29.

@article{
author = { B. Sathya Bama, S. Ramabai, R. Santhanadevi },
title = { Laws based Quality Inspection of Steel Products using Scanning Electron Microscopy (SEM) Images },
journal = { National Conference on Advances in Computer Science and Applications (NCACSA 2012) },
issue_date = { May 2012 },
volume = { NCACSA },
number = { 5 },
month = { May },
year = { 2012 },
issn = 0975-8887,
pages = { 25-29 },
numpages = 5,
url = { /proceedings/ncacsa/number5/6509-1034/ },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
}
%0 Proceeding Article
%1 National Conference on Advances in Computer Science and Applications (NCACSA 2012)
%A B. Sathya Bama
%A S. Ramabai
%A R. Santhanadevi
%T Laws based Quality Inspection of Steel Products using Scanning Electron Microscopy (SEM) Images
%J National Conference on Advances in Computer Science and Applications (NCACSA 2012)
%@ 0975-8887
%V NCACSA
%N 5
%P 25-29
%D 2012
%I International Journal of Computer Applications
Abstract

Since steel is an essential industry raw material and its surface quality is an important evaluation indicator, this paper proposes a quality inspection method for detecting and characterizing defects on steel surfaces. The objective is to detect and classify the defects in the steel products using Scanning Electron Microscopy (SEM) images. In order to obtain better classification accuracy, Discrete Wavelet Transform (DWT) based laws mask method is proposed. Initially, wavelet transform is applied to the input training images and the resultant sub-images are applied with different laws masks like ripple, wave, level, edge and spot. Texture features like mean, entropy, standard deviation, kurtosis and skewness are extracted. The test images are applied with different laws masks and feature values are calculated. These feature values obtained for test and training images are considered for the accuracy assessment which is done based on the minimum distance obtained by taking Sum of Squared Distance (SSD). The accuracy of proposed method is compared with the performance of classical methods namely Tamura features, Gray Level Co-occurrence Matrix (GLCM) and Laws Masks. The overall accuracy of proposed method is 82. 5%. The results obtained indicate that better classification of defects is possible by proposed method of applying DWT based laws masks.

References
  1. Mike Muehlemann, "Standardizing Defect Detection for surface Inspection of Large Web Steel," International Journal of Pattern Recognition and Artificial Intelligence, 2000, Pages: 735-755.
  2. Laurent karsenti, Rehovot, IL. , "Method for defect detection and process monitoring based on SEM images," 2010 US 7,764824 B2.
  3. R. Jeffery Price, W. Kathy Hylton, W. Kenneth Tobin, Jr. , R. Philip Bingham, D. John Hunn and R. John Haines, "Detection of cavitation pits on steel surfaces using SEM imagery," 2008, (865) 574-5743.
  4. I. H. Son, J. D. Lee, S. Choi, D. L. Lee, Y. T Im, "Deformation behavior of the surface defects of low carbon steel in wire rod rolling", Journal of materials processing technology," 2008, 2 0 1pp. 91–96.
  5. Shigeru Takayaa and Kenzo Miyab, "Application of magnetic phenomena to analysis of stress corrosion cracking in welded part of stainless steel,". Journal of Materials Processing Technology 2005, 161 66–74.
  6. Kuldeep Agarwal, Rajiv Shivpuri , Yijun Zhu, Tzyy-Shuh Chang, Howard Huang, "Process knowledge based multi-class support vector classification (PK-MSVM) approach for surface defects in hot rolling,". Expert Systems with Applications 2011, 38 ,7251–7262.
  7. S. G. Mallat, "A theory for multiresolution signal decomposition: the wavelet representation,". IEEE Trans. on pattern analysis and Machine Intelligence, 1989, 11(7): 674~693.
  8. W. Sun, M. Mujherjee, P. Stroeve, A. Palazoglu, J. A. Romagnoli, "A multi-resolution approach for line-edge roughness detection," Microelectronic - 2773 -Eng. 86. 2009, pp. 340-351.
  9. Y. Han, P. Shi, "An adaptive level-selecting wavelet transform for texture defect detection, Image and Vision Computing," Vol. 25, No. 8,2007, pp. 1239-1248.
  10. M. Rachidi, A. Marchadier, C. Gadois, E. Lespessailles, C. Chappard, C. L. Benhamou, "Laws' masks descriptors applied to bone texture analysis: an innovative and discriminant tool in osteoporosis," DOI 10. 1007/s00256-008-0463-2, 2008, 37:541–548.
  11. Tamura and Mori, H. , Yamawaki. Textural Features Corresponding to Visual Perception. IEEE Transaction on Systems, Man, and Cybernetcs, Vol. SMC-8,No. 6, 1978, pp. 460–472.
  12. Hsin-Chih Lin, Chih-Yi Chiu, Shi-Nine Yang, "Finding textures by textual descriptions, visual examples, and relevance feedbacks," Pattern Recognition Letters 24, 2003, pp: 2255–2267. International Journal of Computer Applications (0975 – 8887) Volume *– No. *, ___________ 2012
  13. Haralick, M. Statistical and structural approaches to texture. Proceedings of the IEEE 1978, 67(5): 786–804.
  14. Fritz Albregtsen, "Statistical Texture Measures Computed from Gray Level Coocurrence Matrices," Image Processing Laboratory Department of Informatics University of Oslo, November 5, 2008.
Index Terms

Computer Science
Information Sciences

Keywords

Scanning Electron Microscopy (sem) Feature Extraction Defect Classification Accuracy Assessment