TrackMate-TrackAnalysis extension

I have tried to install the track analysis extension for TrackMate to include

    but it does not appear to recognise the extension, as the track statistics I get after I run Trackmate have not changed; I still get the same output. Is there something I miss?


How exactly did you do the installation?

From what I gather… you need to:

  1. Download the Track Analysis extension
  2. Add that .jar file to your folder
  3. Restart Fiji.

TrackMate will recognize the extra modules and will integrate them in the plugin.

Just try to reinstall it once again… if it’s still not working - report back exactly how you did your installation and the subsequent steps, etc.


Too - make sure to scroll all the way to the end of your “Track statistics” table… when I do the above steps - those extra measures are there.

1 Like

Thank you very much for the attachment. It seems I had downloaded the wrong file from Github. It works fine now.



Hi there,

I’m interested in using this extension - however, from the Github page, I can’t seem to find any definitions for these analysed parameters?

Some of them are self explanatory to me, however, I would like to know how mean direction change rate is defined in this instance.

Do you know where I can find this information?


True, I did not write it yet and sorely miss the time to do so.
You will have to rely on the code in the meantime.

Hello! I’m really sorry to reply on this really old thread! I’m using Trackmate to quantify lots of cell migration experiments - I downloaded the Trackmate track analysis extension, and was hoping to reanalyse my old trackmate files to get the total track length rather than track displacement. Can I load my original trackmate files and get the total track length from these? Or do I need to restart a new analysis each time? Thanks in advance!

Hi Emma

Normally if you use the action called Recompute all features (last panel) the new analyzers should be used.

Fantastic, it works perfectly, thank you so much! This is going to save me a lot of time!

Hello! I am really a new user of Trackmate and this forum. I couldn’t make a post. Thus I just reply to your post.

I want to track water drops (instead of cells). The shapes of the drops are varied a lot (not like cells with similar shapes). Thus, it is very difficult to track the same drop. Thus, I want to use TrackMate-extras to have a try. I downloaded jar file and put it in the jars folder, and restart Fiji. However, I didn’t get what I want. The results are still as same as before.

My working steps are: start Tractmate, select a detector, then next, next, in the end, selected a tracker (from the options, I chose the Linear motion LAP tracker), then start the analysis. My question is: does my working steps right? or I missed some import setting or section?

Thank you very much.

Looking forward your help.

Hello @erdong_yushan

It looks like you are battling with a difficult detection. Can you post an image of your data so that we can see?

Hi, @tinevez

Thank you for your response. I attached three images. One is the raw image, and the other two are the binary images I have processed by using Fiji. At the moment, my approach is only to track the drops in the binary images by the use of trackMate. The drops are located on the right side of the image and flying from right to the left. The size and shapes of the drops are varied which are not like cells. I am wondering whether there is a way to track the drops only by their motion but also by their size, I mean to track the drops with similar size? Thanks!

0002150.tif (2.0 MB) 0002150.tif (2.0 MB) 0002151.tif (2.0 MB)

Ok can you try to operate directly on the raw images, without binarization?
TrackMate does not deal well with B&W images.

Dear @tinevez
Thanks so much for your tips. The raw images work much better than binary images. I want to obtain the drop velocity and its diameter. Thus, I have to get the right track for each droplet.

I would also like to know how does TrackMate calculates the ‘Estimated Diameter’? I checked your previous reply to other questions, you stated that the Estimated Diameter is quite robust. In my results, the estimated diameter is much larger than the '‘real’ diameter (observed by eyes). Thus, I use the particle size obtained from the binary images (Particle analysis of Fiji), then calibrate the results of the Estimated Diameter. Do you think it is a proper way?

Thank you so much for your help!