Re: difference between tiff types

Posted by Daniel James White on Oct 05, 2010; 4:30pm
URL: http://imagej.273.s1.nabble.com/difference-between-tiff-types-tp3686715p3686716.html

Hi  A,
(sorry, cant convert your russian script name to latin!)

On Oct 5, 2010, at 1:40 PM, IMAGEJ automatic digest system wrote:

>
> Date:    Tue, 5 Oct 2010 14:25:05 +0300
> From:    Алексей Малышко <[hidden email]>
> Subject: difference between tiff types
>
> couldn't find it by myself, please could anyone explain.
> i have made some brightness adjustments in ImageJ and saved image (in 16-bit
> grayscale). but when i opened it in irfanview (and other image viewers) i
> saw that image didn't change (like a switched off layer in photoshop). then
> i reopened with imagej and saw changed image. then i converted it in imagej
> to rgb and only after that changes became visible in imageviewers.
>

The pont you are missing is that when you adjust brightness and contrast on a 16 bit tiff image in imageJ
you do not adjust the pixel values, only how they are displayed on the screen (in the 8 bit scale that the screens use)

So if you make and adjustment, then save, the numbers in the image remain unchanged!!!
you only changed how it was being displayed!

BUT...
if you chance brightness and contrast, then convert to 8 bit, it will use the contrast and brightness setting you have applied to also rescale the pixel values to 8 bit.
Then if you open that new  image in another software, it will look how you expect.


you make a very very common mistake.... its simply not clear enough what is going in here in imagej for the novice user!!!

see the imageJ docs here to understand what is going on:
http://rsbweb.nih.gov/ij/docs/menus/image.html#adjust

see especially the auto, set, reset and apply buttons.

and for data types

http://rsbweb.nih.gov/ij/docs/menus/image.html#type



> the question is: what is the practical meaning of

actually thats not the question... but anyway:

> 8–bitgrayscale

image intensities from 0-255
8 bit bit binary representation of that integer range.

> unsigned and signed 16–bitgrayscale
 millions of grey levels


> signed 32–bit floating-point grayscale

a floating point representation of the intensity values using a 32 bit binary number.
http://en.wikipedia.org/wiki/Floating_point

> and
> 32–bit RGB color.

three channels per pixel, each being 8 bit is 24 bit RGB.
32 bit RGB is then .... umm not possible, as you can divide 32 by 3 and get an integer.
I think thats a typo???

anyone explain that?

48 bit RGB is 16 bit per color channel.

>
> or is there a place where i can read about it.
> cannot find it in internets.
> thanks in advance.

Dr. Daniel James White BSc. (Hons.) PhD
Senior Microscopist / Image Visualisation, Processing and Analysis
Light Microscopy and Image Processing Facilities
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstrasse 108
01307 DRESDEN
Germany

[phone] (German Mobile)
[phone] (Work phone at MPI-CBG)
[phone] (Fax MPI-CBG LMF)

http://www.bioimagexd.net  BioImageXD
http://pacific.mpi-cbg.de                Fiji -  is just ImageJ (Batteries Included)
http://www.chalkie.org.uk                Dan's Homepages
https://ifn.mpi-cbg.de  Dresden Imaging Facility Network
dan (at) chalkie.org.uk
( white (at) mpi-cbg.de )