Painting/Drawing in 32-bit images

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Painting/Drawing in 32-bit images

Burger Wilhelm
Hello IJ-list,

is there any way to apply the draw/paint tools on 32-bit (float) images in a useful way? For example, by picking a particular (float) value with the pipette, select a ROI and then fill the region with exactly that value?

It appears that the pipette picks the displayed (lookup) color for the selected pixel but it is not obvious how the corresponding draw/fill value is specified - it is generally not the float value of the selected pixel.

Thanks much for any hints -

Wilhelm



 
Reply | Threaded
Open this post in threaded view
|

Re: Painting/Drawing in 32-bit images

Frederic V. Hessman
Unfortunately, the pipette picks up the displayed value only (i.e. the  
LUT grayscale value for 16- and 32-bit images).

A very GOOD idea to have the pipette pick up (and the filler fill  
with) the right value for the right image (3 values for color, 1  
integer for 8- and 16- bit grayscale, and a double for a 32-bit image)!

Rick

On 2 Jun 2010, at 08:51, Burger Wilhelm wrote:

> Hello IJ-list,
>
> is there any way to apply the draw/paint tools on 32-bit (float)  
> images in a useful way? For example, by picking a particular (float)  
> value with the pipette, select a ROI and then fill the region with  
> exactly that value?
>
> It appears that the pipette picks the displayed (lookup) color for  
> the selected pixel but it is not obvious how the corresponding draw/
> fill value is specified - it is generally not the float value of the  
> selected pixel.
>
> Thanks much for any hints -
>
> Wilhelm
>
>
>
>
Reply | Threaded
Open this post in threaded view
|

Re: Painting/Drawing in 32-bit images

Michael Schmid
In reply to this post by Burger Wilhelm
Hi Wilhelm,

the paint/draw (erase) tools take the foreground (background) color,  
which is an 8-bit value for a grayscale image. It is converted to a  
data value according to the current lookup table. This won't be mor  
accurate than you can have with 8 bits.

For painting black or white, use brightness/contrast and 'Set' to set  
the limits of the display range so that it agrees with the value you  
want to have.
For other values, it is better to create a roi and then use  
Process>Math>Set to set the pixels to any value you like.

Michael
________________________________________________________________

On 2 Jun 2010, at 08:51, Burger Wilhelm wrote:

> Hello IJ-list,
>
> is there any way to apply the draw/paint tools on 32-bit (float)  
> images in a useful way? For example, by picking a particular  
> (float) value with the pipette, select a ROI and then fill the  
> region with exactly that value?
>
> It appears that the pipette picks the displayed (lookup) color for  
> the selected pixel but it is not obvious how the corresponding draw/
> fill value is specified - it is generally not the float value of  
> the selected pixel.
>
> Thanks much for any hints -
>
> Wilhelm
>
>
>
>