[ome-users] displaying histogram to adjust the intensity

[Thomas Julou]

Hi Will,

> The challenge we have for OMERO compared to the other tools you mentioned is that we don?t have all the raw pixel data for the
> image available on the client side.
> 
> You are correct: when you draw a ROI in Insight, we do load all the pixel data for the ROI to construct a histogram and calculate other parameters.
> However, we know that this fails with large images, due to memory issues.
> Also, this only gives you a histogram for the image plane that you are currently viewing, not all the pixels in a multi-plane Z-stack or Time lapse image.

Thanks for spelling out the challenges…
Still, can’t you retrieve histogram values from the server along with the image (or even in the header of the 8-bit image)?

> One approach we have been looking at is to provide an "8-bit histogram", using the rendered image pixel values instead of the raw data.
> This still provides a good impression of the pixel intensity distribution, without loading all the raw data for a plane.
> You can see a prototype of this at:
> http://will-moore.github.io/weblabs/canvas_rendering/ <http://will-moore.github.io/weblabs/canvas_rendering/>

Great! Looking forward to seeing this implemented (with a log-y checkbox please ;)
btw even if you stick to computing the histogram form the 8-bit image on the client side, you should be able to scale the hist using the min/max info that you have anyhow (they’re shown on the current rulers…). Than you could show the cutoff level intensity as well. Does this make sense?

> However, this still only uses a single plane from the image, ignoring other Z-sections and Time points.
> But perhaps this would be enough to provide you with a guide for adjusting rendering settings?

I’m fine with that for the moment. But I realize this is very important to some people… To this matter again, computing the histogram on the server side seems to make more sense.

> You can copy and paste rendering settings between images in OMERO, and can also apply the same rendering
> settings to all images in a dataset or across multiple datasets to compare different experimental conditions. 
> E.g. right-click on an image > Rendering Settings > Copy
> Then select a Dataset (image container) in the Tree, right-click > Rendering Settings > Paste

Thanks, I heavily rely on these features indeed!
btw, I would like to raise again the saturation feature that you skipped in your answer. As soon as there are some bright pixels, e.g. dye clump or camera noise, this is really needed…

Best,
Thomas


>> On 16 Dec 2015, at 12:44, Thomas Julou <thomas.julou at normalesup.org> wrote:
>> 
>> Hello,
>> 
>> I finally start using omero more extensively. One thing I really lack compared to other tools (in particular fiji, icy and micromanager) is the possibility to easily view the intensity histogram of an image. As far as I understand it?s possible only in omero.insight and requires to create a roi for the entire image, correct?
>> 
>> This is particularly needed to adjust the contrast of a dataset and/or to compare different experimental conditions.
>> I?d love to see the histogram displayed in the preview pane of omero web (e.g. one coloured line per displayed channel, with a checkbox for log y-axis). Along the same line having a field to set saturation next to the min/max button would be really convenient (similar to the ignore field in micromanager, or to the Auto button in the contrast window of imagej).
>> 
>> Please let me know if these features are already implemented and I wasn?t able to find them. Otherwise, I?d really like to know whether this is on your roadmap. Best regards,
>> 
>> --
>> Thomas Julou  |  Computational & Systems Biology  |  Biozentrum ? University of Basel  |  Klingelbergstrasse 50/70 CH-4056 Basel  |  +41 (0)61 267 16 21
>> 
>> _______________________________________________
>> ome-users mailing list
>> ome-users at lists.openmicroscopy.org.uk
>> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users
> 
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20151217/1a2ebef8/attachment-0001.html>
> 
> ------------------------------
> 
> Subject: Digest Footer
> 
> _______________________________________________
> ome-users mailing list
> ome-users at lists.openmicroscopy.org.uk
> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users
> 
> 
> ------------------------------
> 
> End of ome-users Digest, Vol 129, Issue 8
> *****************************************

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20151217/24024208/attachment.html>