Classification of protein crystallisation images using Fourier descriptors.

Christopher G. Walker, James Foadi, Julie Wilson*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


The two-dimensional Fourier Transform (2D-FT) is well suited to the extraction of features to differentiate image texture and the classification of images based on information acquired from the frequency domain provides a complementary method to approaches based within the spatial domain. The intensity, I, of the Fourier transformed images can be modeled by an equation of power law form, I = Ar?, where A and ? are constants and r is the radial spatial frequency. The power law is fitted over annuli, centred at zero spatial frequency, and the parameters, A and ?, determined for each spatial frequency range. The variation of the fitted parameters across wedges of fixed polar angle provides a measure of directionality and the deviation from the fitted model can be exploited for classification. The classification results are combined with an existing method to classify individual objects within the crystallisation drop to obtain an improved overall classification rate.
Original languageEnglish
Pages (from-to)418-426
Number of pages9
JournalJournal of Applied Crystallography
Issue number3
Publication statusPublished - Jun 2007


  • Classification
  • Crystallization images
  • Fourier transform
  • Pattern recognition

Cite this