Color Deconvolution and Support Vector Machines

  • Charles E. H. Berger
  • Cor J. Veenman
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5718)


Methods for machine learning (support vector machines) and image processing (color deconvolution) are combined in this paper for the purpose of separating colors in images of documents. After determining the background color, samples from the image that are representative of the colors to be separated are mapped to a feature space. Given the clusters of samples of either color the support vector machine (SVM) method is used to find an optimal separating line between the clusters in feature space. Deconvolution image processing parameters are determined from the separating line. A number of examples of applications in forensic casework are presented.


forensic document analysis questioned documents color separation color deconvolution image processing support vector machines 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995)CrossRefzbMATHGoogle Scholar
  2. 2.
    Kecman, V.: Learning and soft computing: support vector machines, neural networks, and fuzzy logic models. MIT Press, Cambridge (2001)zbMATHGoogle Scholar
  3. 3.
    Chen, H.S., Meng, H.H., Cheng, K.C.: A survey of methods used for the identification and characterization of inks. Forensic Science Journal 1, 1–14 (2002)Google Scholar
  4. 4.
    Berger, C.E.H., de Koeijer, J.A., Glas, W., Madhuizen, H.T.: Color Separation in Forensic Image Processing. Journal of Forensic Sciences 51, 100–102 (2006)CrossRefGoogle Scholar
  5. 5.
    Berger, C.E.H.: Inference of identity of source using univariate and bivariate methods. Science and Justice (2009) (in Press) DOI:10.1016/j.scijus.2009.03.003Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Charles E. H. Berger
    • 1
  • Cor J. Veenman
    • 1
    • 2
  1. 1.Netherlands Forensic InstituteThe HagueThe Netherlands
  2. 2.Intelligent Systems LabUniversity of AmsterdamAmsterdamThe Netherlands

Personalised recommendations