I have a series of 37 b/w microscopy photos. The whole sample shifts slightly (by ~10 pixels) from photo to photo (systematic error). See an example photo here and download two photos pos4_0.dat and pos4_40.dat, shifted by ~10px. Moreover, there are major intensity variations. Nevertheless, the main structure is always visible.


I want to track the trajectory of some points through the photo series. Later on, I will stabilize my regions of interest. The last example in Documentation would be excactly what I need, but ImageFeatureTrack doesn't work.

import = Import[#] & /@ {"pos4_0.dat", "pos4_40.dat"}
images = Image[Rescale[#]] & /@ import
Out[55]={{}, {}} 

How can I improve the performance of ImageFeatureTrack or solve my problem in another way?



closed as off-topic by Dr. belisarius, user9660, MarcoB, Jens, m_goldberg Feb 12 '16 at 21:22

  • The question does not concern the technical computing software Mathematica by Wolfram Research. Please see the help center to find out about the topics that can be asked here.
If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    $\begingroup$ Can you post the images? $\endgroup$ – M.R. Sep 30 '15 at 14:21
  • $\begingroup$ You could try helping it out by giving it the coordinates of the feature in the first image. It looks like it's not finding a feature to begin with. $\endgroup$ – N.J.Evans Sep 30 '15 at 14:23
  • $\begingroup$ @M.R. raw data is given as download link in the first paragraph. Does this work for you? $\endgroup$ – keyx Sep 30 '15 at 14:27
  • $\begingroup$ Your Images are 8 MB each. Perhaps you could reduce the resolution for this question ... $\endgroup$ – Dr. belisarius Sep 30 '15 at 14:36
  • 4
    $\begingroup$ I'm voting to close this question as off-topic because the sample images linked in the question aren't available for download and the question is not answerable without them $\endgroup$ – Dr. belisarius Feb 12 '16 at 14:17

Browse other questions tagged or ask your own question.