🚀 go-pugleaf

spl@ivem.ucsd.edu (Steve Lamont) writes:

>In article <fred.722416637@dip3> fred@dip3.ee.uct.ac.za (Fred Hoare) writes:
>>I (and the other people in my lab) have a requirement to be able
>>to display 16 bit greyscale images on a Sun workstation. We have
>>several pieces of software that display images, but they all convert
>>the images to 8 bit first. What we require is a display routine that
>>would allow a person to choose what intensity range in the image was
>>displayed on the screen, at any time. I presume this would be done by 
>>changing the way the lookup table gets mapped to the data.

>The way that I attack it is to either just truncate the lower 8 bits
>or to scale the data so that the maximum value is mapped into 255 and
>the minimum value is mapped into 0.  The can either be done by table
>lookup or through some direct scaling.

>I'm not sure what is meant by "allow a person to choose what intensity
>range was displayed on the screen..."  I suppose that you could use a
>sliding eight bit window that could be used as a mask to map into the
>color table but without a lot of fiddling the images are going to look
>awfully strange.

That is exactly what I meant. The images we work with consist of both
medical and satelite images and it is often very helpful to be able to
look at a specific range of intensities in the image. It is a lot easier
to do this using a lookup table modification than to actually truncate
the image to the range required, then scale this to 8 bits before display.

Fred
--
Frederick Hoare                email: fred@dip1.ee.uct.ac.za
Image Processing Laboratory
Department of Electrical and Electronic Engineering
University of Cape Town,  South Africa