grnidone - 3:30 am on Mar 24, 2001 (gmt 0)

Tedster,

I have tried the technique over and over again and in every case the image with the a and b sections with a gausion blur are *bigger*.

I am convinced this technique will work, but I'll be damned if I can figure out how to make it work. In some cases, the 'treated' image is twice the size of the untreated image.

My process:

I take an image from a photo cd..something where I know the image will be very sharp and clear. It is usually a jpg file. I immediately save it as a png file...that is my "control".

Then I go back to the same image on the cd and open it up and change it to an LAB. I do the gaussion blur on the a and be sections, turn the photo back to a RGB and save the mess as a png file to compare to the control.

Every time the LAB treated photo is bigger. Every time.

Is this an E.S.T.O. problem?

-G

And of course, everyone knows that E.S.T.O stands for "Equipment Superior To Operator"