Skip to main content
SpringerLink
Log in

Robust automated multiple view inspection

  • Theoretical Advances
  • Published:
Pattern Analysis and Applications Aims and scope Submit manuscript

Abstract

Recently, Automated Multiple View Inspection (AMVI) has been developed for automated defect detection of manufactured objects, and the framework was successfully implemented for calibrated image sequences. However, it is not easy to be implemented in industrial environments because the calibration is a difficult and an unstable process. To overcome these disadvantages, the robust AMVI strategy, which assumes that an unknown affine transformation exists between each pair of uncalibrated images, is proposed. This transformation is estimated using two complementary robust procedures: a global approximation of the affine mapping is computed by creating candidate correspondences via B-splines and selecting those which better satisfy the epipolar constraint for uncalibrated images. Then, we use this approximation as initial estimate of a robust intensity-based matching approach, which is applied locally on each potential defect. The result is that false alarms are discarded, and the defects of an industrial object are actually tracked along the uncalibrated image sequence. The method is successful as shown in our experiments on aluminum die castings.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Notes

  1. For instance in printed circuit board (PCB) inspection.

  2. In this paper we use affine transformations, although it is also possible to implement perspective transformations.

  3. Occlusions appear when small flaws move in front (or behind) of a thick cross section of the casting, where X-rays are highly absorbed; and when flaws are located in the outer limits of the visible area of the casting.

  4. Digital radioscopic images are generated using a frame grabber, which averages n samples of the scene taken at infinitesimal time intervals in order to reduce noise and improve the signal-to-noise ratio.

  5. B-splines are invariant under affine transformations. In practice, linear splines can also be utilised with enough number of knots.

  6. Do not confuse the fundamental matrix F with the affine mapping H.

  7. Alternatives to choose ρ, for instance, are: Cauchy, Huber, Tukey, Geman-McClure and Lorentzian robust functions [27]. In our experiments we use the Geman-McClure one.

  8. Inspection approaches which make use of only one view are also affected by this problem.

References

  1. Mery D, Filbert D (2002) Automated flaw detection in aluminum castings based on the tracking of potential defects in a radioscopic image sequence. IEEE Trans Rob Autom 18(6):890–901

    Article  Google Scholar 

  2. Mery D, Carrasco M (2005) Automated multiple view inspection based on uncalibrated image sequences. Lect Notes Comput Sci 3540:1238–1247

    Article  Google Scholar 

  3. Abramovich G, Barhak J, Spicer P (2005) Reconfigurable array for machine vision inspection (RAMVI). In: Proceedings of the 3rd international conference on reconfigurable manufacturing, Ann Arbor

  4. Spicer P, Bohl K, Abramovich G, Barhak J (2006) Robust calibration of a reconfigurable camera array for machine vision inspection (RAMVI) using rule-based colour recognition. In: Proceedings of the 1st international conference on computer vision theory and applications, Setúbal, pp 131–138

  5. Gumustekin S (2004) A visual inspection system using a single camera and mirrors. In: Proceedings of the 12th IEEE signal processing and communications applications conference, Kusadasi, pp 257–260

  6. Hartley RI, Zisserman A (2002) Multiple view geometry in computer vision. Cambridge University Press, Cambridge

    Google Scholar 

  7. Boerner H, Strecker H (1988) Automated X-ray inspection of aluminum casting. IEEE Trans Pattern Anal Mach Intell 10(1):79–91

    Article  Google Scholar 

  8. Filbert D, Klatte R, Heinrich W, Purschke M (1987) Computer aided inspection of castings. In: IEEE-IAS Annual Meeting, Atlanta, USA, pp 1087–1095

  9. Heinrich W (1988) Automatische Röntgenserienprüfung von Gußteilen. PhD Thesis, Institut für Allgemeine Elektrotechnik, Technische Universität Berlin

  10. Mohr G, Fock T (2004) X-ray inspection in the aerospace industry—state of the art, challenges and emerging techonologies. In: Proceedings of the 16th world conference on nondestructive testing, Montreal

  11. Mahi Uddin Khan Md (1998) Non-destructive testing applications in commercial aircraft maintenance. In: Proceedings of the 7th European conference on non-destructive testing, Copenhagen

  12. Kita Y, Highnam R, Brady M (2001) Correspondence between different view breast X-rays using curved epipolar lines. Comput Vis Image Underst 83(1):38–56

    Article  MATH  Google Scholar 

  13. Rebuffel V, Pires S, Caplier A, Lamarque P (2003) Automatic delamination defects detection in radiographic sequences of rocket boosters. In: Proceedings of the international symposium on computed tomography and image processing for industrial radiology, Berlin

  14. Jähne B, Haußecker H, Geißler P (1999) Handbook of computer vision and applications, vol 2. Signal processing and pattern recognition. Academic, San Diego

    Google Scholar 

  15. Mery D (2003) Crossing line profile: a new approach to detecting defects in aluminium castings. Lect Notes Comput Sci 2749:725–732

    Article  Google Scholar 

  16. Mery D, da Silva R, Caloba LP, Rebello JMA (2003) Pattern recognition in the automatic inspection of aluminium castings. Insight 45(7):475–483

    Article  Google Scholar 

  17. Haralick RM, Shapiro LG (1992) Computer and robot vision. Addison-Wesley, New York

    Google Scholar 

  18. Sonka M, Hlavac V, Boyle R (1998) Image processing, analysis, and machine vision, 2nd edn. PWS Publishing, Pacific Grove

    Google Scholar 

  19. de Boor C (1978) A practical guide to splines. Springer, New York

    MATH  Google Scholar 

  20. Sampson PD (1982) Fitting conic sections to ‘very scattered’ data: an iterative refinement of the bookstein algorithm. Comput Vis Graph Image Process 18:97–108

    Article  Google Scholar 

  21. Wells WM III (1997) Statistical approaches to feature-based object recognition. Int J Comput Vis 21(1–2):63–98

    Article  Google Scholar 

  22. Hsieh JW, Liao HYM, Fan KC, Ko MT, Hung YP (1997) Image registration using a new edge-based approach. Comput Vis Image Underst 67:112–130

    Article  Google Scholar 

  23. Black MJ, Jepson AD (1998) Eigen-tracking: robust matching and tracking of articulated objects using a view-based representation. Int J Comput Vis 26(1):63–84

    Article  Google Scholar 

  24. Hager GD, Belhumeur PN (1998) Efficient region tracking with parametric models of geometry and illumination. IEEE Trans Pattern Anal Mach Intell 20(10):1025–1039

    Article  Google Scholar 

  25. Baker S, Matthews I (2004) Lukas-Kanade 20 years on: a unifying framework. Int J Comput Vis 56(3):221–255

    Article  Google Scholar 

  26. Lucas BD, Kanade T (1981) An iterative image registration technique with an application to stereo vision. In: Proceedings of the 7th international joint conference on artificial intelligence, Vancouver, pp 674–679

  27. Black MJ, Anandan P (1996) The robust estimation of multiple motions: parametric and piecewise-smooth flow fields. Comput Vis Image Underst 63(1):75–104

    Article  Google Scholar 

  28. Stewart CV (1999) Robust parameter estimation in computer vision. SIAM Rev 41(3):513–537

    Article  MATH  MathSciNet  Google Scholar 

  29. Rousseeuw PJ, Leroy AM (1987) Robust regression and outlier detection. Wiley, New York

    MATH  Google Scholar 

  30. Holland PW, Welsch RE (1977) Robust regression using iteratively reweighted least-squares. SIAM Rev 6:813–827

    Google Scholar 

  31. Baker S, Matthews I (2001) Equivalence and efficiency of image alignment algorithms. In: Proceedings of the 8th IEEE conference on computer vision and pattern recognition, Vancouver, vol 1, pp 1090–1097

  32. Bruhn A, Weickert J, Kohlberger T, Schnörr C (2006) A multigrid platform for real-time motion computation with discontinuity-preserving variational methods. Int J Comput Vis 70(3):257–277

    Article  Google Scholar 

Download references

Acknowledgments

This work was supported by FONDECYT—Chile under grant no. 1040210.

Author information

Authors and Affiliations

  1. Mathematical Image Analysis Group, Faculty of Mathematics and Computer Science, Saarland University, Building E11, 66041, Saarbrücken, Germany

    Luis Pizarro

  2. Escuela de Ingeniería Informática, Facultad de Ingeniería, Universidad Diego Portales, Av. Ejército 441, Santiago, Chile

    Luis Pizarro

  3. Departamento de Ciencia de la Computación, Pontificia Universidad Católica de Chile, Av. Vicuña Mackenna, 4860, Santiago, Chile

    Domingo Mery, Rafael Delpiano & Miguel Carrasco

Authors
  1. Luis Pizarro
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Domingo Mery
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Rafael Delpiano
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Miguel Carrasco
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Luis Pizarro.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Pizarro, L., Mery, D., Delpiano, R. et al. Robust automated multiple view inspection. Pattern Anal Applic 11, 21–32 (2008). https://doi.org/10.1007/s10044-007-0075-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10044-007-0075-9

Keywords

Search

Navigation