Re: 16-bit raw images - how to preserve absolute value?

Posted by Michael Schmid on Apr 08, 2010; 2:02pm
URL: http://imagej.273.s1.nabble.com/16-bit-raw-images-how-to-preserve-absolute-value-tp3688645p3688646.html

Hi Robert,

it seems to be either a problem of incorrectly giving little endian/
big endian byte order or the offset to the first image is wrong in  
the import dialog (it should be 0 for raw files written by ImageJ)

Edit>Options>IO Options allows you to set whether to write raw files  
in Intel (=little endian) format.

Michael
________________________________________________________________

On 8 Apr 2010, at 12:21, Robert Atwood wrote:

> Hi;
> I've been using ImageJ to prepare data files for some of my own
> routines, because I can create shapes interactively in the  
> interface or
> by macro. However, now when I try to load a file which is 16-bit raw
> data, containing values that are only small integers (0,1,2,3) , then
> change (for example ) areas of '0' to '1' using flood-fill, then  
> save as
> raw data, then import it again, I find that the values are rescaled  
> such
> that 1 becomes 256, 2 becomes 512, and 3 becomes 768
>
> How do I avoid this ? I need the values to remain exactly as I assign
> them!
>
>  Thanks
> Robert