> #1. Is it possible to have the outputted filename derived from the  
> original image's filename? For example, if you are selecting from an  
> image named "somegirl.gif" then could the script assume that the  
> output filename should be named "somegirl-cutout.gif"?

No. Picture 2 is derived from the page number  where it is,
for example the first image on page 3248 of picture 100.gif is called 
32481.gif, picture2 32482.gif, etc...
page number is contained by the original picture, and has nothing to do
with it's name. Original picture is called  in the example 100.gif, and the 
derived pictures are called 32481.gif,  32482.gif and 32491.gif
To find the proper name is job for a human, the machine
can not effectively help there.

> #2. Is it necessary to create a new image from the copied region? And  
> even if it is (e.g., you want to edit the pasted image), would it be  
> acceptable if the script performed the creation of this image. I.e.,  
> you make your selection and then run the script -- the new image is  
> opened (if necessary) and then saved as a GIF.

The script is unable to create the images. They must be created manually.
They are pictures in written text, and only a human can decide,
which parts are images and which parts are text. The selected
part then usually gets modified in size and also sometimes rotated.
All jobs, only a human can do properly.
I checked with the most up-to-date ocr programs, they do a lousy
job when they try to find out, which parts are pictures and
which are text. They err in 20-30% of the cases, which 
is completely unacceptable and unusable for me.

> Of course, doing precisely what you described is possible; it just  
> entails some steps that might better be automated.

That exactly would speed up my work by at least 20-30%.

Thanks, eleonora

GMX startet ShortView.de. Hier findest Du Leute mit Deinen Interessen!
Jetzt dabei sein: http://www.shortview.de/[EMAIL PROTECTED]
Gimp-user mailing list

Reply via email to