Skip to content

KerdenSOMClassification

Jump to bottom
Adrian Quintana edited this page Dec 11, 2017 · 1 revision

2D KerdenSOM Classification

  • Graphically design a mask that defines the region of interest
  • Clasify using a self-organizing map
  • Visualize the self-organizing map, and identify distinct classes

Example Data

  • Input
    • Gallery of images. Previous alignment made with maximum likelihood is assumed.

File kerdenSOM_classification.tar.gz contains two sets of images and the xmipp_protocol_kerdensom.py script with suitable parameters.

  • Output
    • One selection file per class. Each selection file contains the images assigned to each class
    • Auxiliary files used by the analyze results.

File output_kerdenSOM_classification.tar.gz contains the results of executing theKerdenSOM script.

Self organizing map and average images of the two selected classes
Press`analyze results` button for visualizing the classification.

NOTE: since this protocol uses random seeds, the results may vary between different runs.

USER's COMMENTS

--Main.RobertoMarabini - 09 Oct 2007

Clone this wiki locally