Extraction of Local Structural Features in Images by Using a Multi-scale Relevance Function

  • Roman M. Palenichka
  • Maxim A. Volgin
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1715)


Extraction of structural features in radiographic images is considered in the context of flaw detection with application to industrial and medical diagnostics. The known approache, like the histogram-based binarization yield poor detection results for such images, which contain small and low-contrast objects of interest on noisy background. In the presented model-based method, the detection of objects of interest is considered as a consecutive and hierarchical extraction of structural features (primitive patterns) which compose these objects in the form of aggregation of primitive patterns. The concept of relevance function is introduced in order to perform a quick location and identification of primitive patterns by using the binarization of regions of attention. The proposed feature extraction method has been tested on radiographic images in application to defect detection of weld joints and extraction of blood vessels in angiography.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    A. Kehoe and G. A. Parker. Image processing for industrial radiographic inspection: image inspection, British Journal of NDT, Vol. 32, N. 4, pp. 183–190, 1990.Google Scholar
  2. 2.
    M. Sonka et al. Adaptive approach to accurate analysis of small-diameter vessels in cineangiograms, IEEE Trans. on Medical Imaging, Vol. 16, No. 1, pp. 87–95, 1997.CrossRefGoogle Scholar
  3. 3.
    J. J. Heine et al. Multiresolution statistical analysis of high-resolution digital mammograms, IEEE Trans. on Medical Imaging, Vol. 16, No. 5, pp. 503–515, 1997.CrossRefGoogle Scholar
  4. 4.
    A. K. Jain and M.­P. Dubuisson. Segmentation of X­ray and C­scan images of fiber reinforced composite materials, Pattern Recognition, Vol. 25, No. 3, pp. 257–270, 1992.CrossRefGoogle Scholar
  5. 5.
    T. W. Liao and J. Ni. An automated radiographic NDT system for weld inspection: Part I ­ Weld extraction, NDT&EInternational, Vol. 29, No. 3, pp. 157–162, 1996.Google Scholar
  6. 6.
    T. W. Liao and Y. Li. An automated radiographic NDT system for weld inspection: Part II-Flaw detection, NDT&EInternational, Vol. 31, No. 3, pp.183–192, 1998.MathSciNetGoogle Scholar
  7. 7.
    S. Mallat. A theory for multi-resolution image signal decomposition: the wavelet representation, IEEE Trans., Vol. PAMI­it11, pp. 674–693, 1989.Google Scholar
  8. 8.
    Ya. Chitti. Detection of small intensity changes in CCD images with non-uniform illumination and large signal dependent noise, CVGIP: Graph. Models and Image Processing, Vol. 59, No. 3, pp. 139–148, 1997.CrossRefGoogle Scholar
  9. 9.
    C.H. Chen and G.G. Lee. On digital mammogram segmentation and microcalcification detection using multi-resolution wavelet analysis, CVGIP: Graph. Models and Image Processing, Vol. 59, No. 5, pp.348–364, 1997.MathSciNetGoogle Scholar
  10. 10.
    K. Fukunaga. Introduction to Statistical Pattern Recognition, Second Edition, New York: Academic Press, 1990.zbMATHGoogle Scholar
  11. 11.
    K. Kanatani. Hypothesizing and testing geometric properties of image data, CVGIP: Image Understanding Vol. 54, pp. 349–357, 1992.CrossRefGoogle Scholar
  12. 12.
    M. D. Wheeler and K. Ikeuchi. Sensor modeling, probabilistic hypothesis generation, and robust localization for object recognition, IEEE Trans. Vol. PAMI-17, No. 3, pp. 252–266, 1995.Google Scholar
  13. 13.
    A. B. Frakt, W.C. Karl and A. S. Willsky. A multiscale hypothesis testing approach to anomaly detection and localization from noisy tomographic data, IEEE Trans. Image Proc., Vol. 7, No. 6, pp. 404–420, 1998.Google Scholar
  14. 14.
    R. M. Palenichka and U. Zscherpel. Detection of flaws in radiographic images by hypothesis generation and testing approach, Proc. Int. Workshop Advances in Signal Processing for NDE of Materials, pp. 235–243, 1998.Google Scholar
  15. 15.
    A. Rosenfeld and S. Banerjee. Maximum­likelihood edge detection in digital signals, CVGIP: Image Understanding, Vol. 55, No. 1, 1992.Google Scholar
  16. 16.
    L.P. Kaebling and A. W. Moore. Reinforcement learning: a survey, Journal of Artificial Intelligence Research, No. 4, pp. 237–285, 1996.Google Scholar
  17. 17.
    P. Perner, T. P. Belikova and N. I. Yashunskaya. Knowledge acquisition by symbolic decision tree induction for interpretation of digital images in radiology, Proc. Int. Workshop Structural and Syntactic Pattern Recognition, pp. 208–220, 1996.Google Scholar
  18. 18.
    S. Katsuragawa et al. Classification of normal and abnormal lungs with interstatial diseases by rule-based methods and artificial neural networks, Journal of Digital Imaging, vol. 10, No. 3, pp. 108–114, 1997.CrossRefGoogle Scholar
  19. 19.
    P. Perner. A knowledge-based image inspection system for automatic defect recognition, classification, and process diagnosis,Machine Vision and Applications, No. 7, pp. 135–147,1994.CrossRefGoogle Scholar
  20. 20.
    P. Perner and W. Paetzold. An incremental learning system for interpretation of images, Proc. Int. Workshop Shape, Structure and Pattern Recognition, pp. 311–319, 1994.Google Scholar
  21. 21.
    R. M. Palenichka et al. Structure-adaptive image filtering using order statistics, Journal of Electronic Imaging, No. 2, pp. 339–349, 1998.CrossRefGoogle Scholar
  22. 22.
    L. Itti, C. Koch, and E. Niebur. A model of saliency-based visual attention for rapid scene analysis, IEEE Trans., Vol. PAMI­ 20, No. 11, pp. 1254–1259, 1998.Google Scholar
  23. 23.
    T. C. Folsom and R. B. Pinter. Primitive features by steering, quadrature, and scale, IEEE Trans., Vol. PAMI­ 20, No. 11, pp. 1161–1173, 1998.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Roman M. Palenichka
    • 1
  • Maxim A. Volgin
    • 1
  1. 1.Inst. of Physics & MechanicsLvivUkraine

Personalised recommendations