Login  Register

Incorrect density values when quantifying PNG files exported from GIMP

Posted by Garvalov, Boyan on Nov 19, 2012; 11:24am
URL: http://imagej.273.s1.nabble.com/Incorrect-density-values-when-quantifying-PNG-files-exported-from-GIMP-tp5000862.html

Hello!

 

I have been using ImageJ to quantify some images (western blot scans, grayscale). The blots were first scanned in JPG format, then were rotated and cropped using GIMP and finally exported from GIMP either in JPG or PNG format.

When quantifying the images in ImageJ (version 1.47f) I realized that the images exported in JPG format give essentially the same mean and integrated density (IntDen/RawIntDen) values as the corresponding area of the original scan, however, the images exported in PNG format give very different values (depending on the image - up to >2-fold differences in IntDen). The PNG Image also looks visually quite different from the JPG image and the original scan when viewed in ImageJ, but not in a number of other viewers.

This problem does not occur if I export from GIMP to TIF, or if I use Photoshop to save as PNG. If I "quantify" the images in either GIMP or Photoshop using the histogram function I get identical mean values for all image formats, corresponding to the mean value for the JPG file that ImageJ gives me.

When exporting to PNG format from GIMP I had all the options unchecked except the following: 1) save resolution, 2) save creation time, 3) save comment, 4) compression level was set to the default value of  9; 5) I tried with the "save background colour" option checked or unchecked by that didn't make a difference for the values I obtained in ImageJ.

I realize that this may be a tricky question as it refers to a very specific issue at the interface of two pieces of software and that there are obvious workarounds. However, I do find GIMP more convenient for processing the original scans and the PNG format has some advantages over JPG, so any hints as to how the issue may be solved using GIMP-generated PNG format would be appreciated!

 

Thanks

Boyan

 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Boyan K. Garvalov
Institut für Neuropathologie
Justus-Liebig-Universität
Aulweg 123, 3. OG
35392 Giessen
 
Tel: (+49-641) 99-41401
Fax: (+49-641) 99-41689
E-mail: [hidden email] <mailto:[hidden email]>
Website: http://www.ukgm.de/ugm_2/deu/ugi_npa/5424.html <http://www.ukgm.de/ugm_2/deu/5424.html>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html