Thread Counting in X-Rays of Plain-Weave Painting Canvas

  • Juan José Murillo-FuentesEmail author
  • Laura Alba


In this work we present some results on the development and test of software for thread counting in X-rays images of plain-weave canvas. We have reviewed the state of the art in this area to select good and simple approaches to be implemented. Hence, we propose the fast Fourier transform (FFT) as the core of our software. We show how with little modifications of the software the frequency of line patterns in simple twills and zig-zag twills can be also analyzed. The algorithm looks for most probable and mean values within a predefined range of frequency values. The method provides as output a couple of images with the local frequency in every point of the selected area. It also includes the estimated degree deviation from the vertical and horizontal for vertical and horizontal lines (threads in simple-weave) for every point, as images. The histograms of the frequencies found in the X-ray are also displayed. The software was designed following the requirements of the Museo Nacional del Prado Conservation and Research department. It was tested with X-rays authored by 17th century painters Miguel de Pret, Claudio de Lorena, Van Dyck and Rubens. In the analysis of a single image, the results of the software have proved useful in the automatic frequency estimation, detection of cusping and analysis of cut edges or areas. In the two images comparison, the method has been already used to draw conclusions on the canvas coming from the same roll, based on the thread counting. Furthermore, we propose the analysis of the histogram as a useful feature to help in the comparison.


  1. 1.
    Barni M, Pelagotti A, Piva A (2005) Image processing for the analysis and conservation of paintings: opportunities and challenges. IEEE Sig Process Mag 22(5):141–144. Scholar
  2. 2.
    Johnson CR Jr, Hendriks E, Berezhnoy IJ, Brevdo E, Hughes SM, Daubechies I, Li J, Postma E, Wang JZ (2008) Image processing for artist identification. IEEE Sig Process Mag 25(4):37–48. Scholar
  3. 3.
    Yang H, Lu J, Brown WP, Daubechies I, Ying L (2015) Quantitative canvas weave analysis using 2-D synchrosqueezed transforms. IEEE Sig Process Mag 32(4):55–63. Scholar
  4. 4.
    van der Maaten L, Erdmann R (2015) Automatic thread-level canvas analysis. IEEE Sig Process Mag 32(4):38–45. Scholar
  5. 5.
    Johnson CR Jr, Johnson DH, Verslype I, Lugtigheid R, Erdmann RG (2013) Detecting weft snakes. ArtMatters Int J Tech Art Hist 5:48–52Google Scholar
  6. 6.
    Fondón-García I, Simois FJ, Murillo-Fuentes JJ (2014) Software tool for thread counting in X-rays of plain-weave painting canvas. In: Proceedings of 11th international conference on non-destructive investigations and microanalysis for diagnostics and conservation of cultural and environmental heritageGoogle Scholar
  7. 7.
    Erdmann R, Johnson CR, Jr, Schafer M, Twilley J (2013) Reuniting Poussin’s Bacchanals painted for Cardinal Richelieu through quantitative canvas weave analysis. In: Proceedings of 41st annual meeting American institute conservation historic and artistic worksGoogle Scholar
  8. 8.
    Escofet J, Millán MS, Ralló M (2001) Modeling of woven fabric structures based on Fourier image analysis. Appl Opt 40:6170–6176CrossRefGoogle Scholar
  9. 9.
    Johnson DH, Sun L, Johnson CR, Hendriks E (2010) Matching canvas weave patterns from processing X-ray images of master paintings. In: IEEE international conference on acoustics speech and signal processing (ICASSP), pp 958–961Google Scholar
  10. 10.
    Oppenheim AV, Schafer RW (1999) Discrete-time signal processing. Prentice-Hall, Upper Saddle River, NJGoogle Scholar
  11. 11.
    Osgood B (2016) Lecture notes for EE261, the Fourier Transform and its applications, 3rd ed,, Cambridge Stanford University. Accessed 19 Nov 2016
  12. 12.
    Johnson DH, Hendriks E, Geldof M, Johnson CR, Jr (2010) Do weave matches imply canvas roll matches? In: Proceedings of 38th annual meeting American institute conservation historic and artistic worksGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Dep. Teoría de la Señal y ComunicacionesEscuela Técnica Superior de Ingeniería. Universidad de SevillaSevilleSpain
  2. 2.Documentación Técnica y LaboratorioMuseo Nacional del PradoMadridSpain

Personalised recommendations