I responded but the message seems lost. In short:
The person has design ready somewhere so he should be able to get highres images. Using clever scaling of some programs can lead to better image. However eyes work well on the interpolation. And if the original image exists somehow getting the highres image should be rather easy.

A bad res image would be like this:

http://i378.photobucket.com/albums/oo221/Toni_Vakiparta/misc/Freeship_backroundimage1.jpg

and a good one like this:
http://i378.photobucket.com/albums/oo221/Toni_Vakiparta/misc/Freeship_backroundimage2.jpg

: The original image may remain unchanged, but consider what happens when the
: Delftship (or any other image-manipulation program) needs to display an
: image in a larger size.
: Unlike say, images painted on rubber, images made of pixels do not have any
: inherent "stretch" ability.
: In order to show pixel-based image in a larger size, a new image must be made
: with greater number of pixels in it. Depending on the program, the new
: image may be only temporary (exist only in program memory) which may be
: the case with Delftship.
: Pixel is only a small rectangular point on the screen that is fixed in size,
: so the larger the image, the more pixels it needs.
: Let's say that you have a black curving line on a white background. If you
: enlarged the image you'd see the line's edges
: are not smooth but resemble "steps". If a human was in charge of
: enlarging the black line, he would simply add more black
: pixels only.
: However, software does not know what the pixels represent.
: Thus, it must make some educated guess what pixels of what color to add.
: Depending on the algorithm it may do better or worse job in that guessing.
: Now, black line on white background is a very simple pattern. In color
: photographs the pixel structure of the objects represented may be very
: complex, thus making the software interpolation a very challenging task.