|
I spent some time dealing with this problem today and I couldn't figure out how to change the anti-aliasing (I'm using Inkscape, maybe there's a way to do it?). Eventually I stumbled on a simple but not very elegant solution:
1. open the .eps file
2. ungroup everything: after 3-4 ungroups I had it down to a layer in which all of the triangles can be selected with a single click.
3. copy the triangles and paste in place a couple times. The lines get fainter and disappear eventually.
"Vinesh Rajpaul" wrote in message <i1g11s$jt9$1@fred.mathworks.com>...
> Hi -
>
> The problem you mentioned (the "white crisscrossed lines") frustrated the hell out of me for a while. As far as I can tell, the problem lies not with MATLAB's image rendering but rather with the anti-aliasing routine used in most PDF/EPS viewers. In other words the output from MATLAB is fine but the PDF viewer is not rendering the image correctly. If you can find a way to turn the anti-aliasing (in your PDF/EPS viewer) off, I think the problem will likely go away. E.g. if you are using Adobe Reader: go to Preferences > Page Display > disable the "smooth line art" setting. This sorted out the issue for me!
>
> "Dan Jones" <Daniel.Jones@colostate.edu> wrote in message <hqpv1p$hi$1@fred.mathworks.com>...
> > Hi all,
> >
> > I'm using pcolor (with interpolated shading) to create a figure. My goal is to then use Adobe Illustrator (CS3) to produce a publication-quality graphic.
> >
> > The figure that Matlab produces looks great, but once I save the image in EPS or PDF format, the quality goes down considerably. The image is crisscrossed with white lines that form triangular polygons. I guess those are the primitive objects that were used to draw these particular vector graphic images. Unfortunately, it looks awful.
> >
> > I've tried saving the Matlab output in Illustrator format and then opening this file in illustrator, but the results are terrible - the image looks very blocky. Saving the Matlab output as a bitmap doesn't seem to work, either. The quality is too low.
> >
> > Is this just a data resolution problem? That is, should I try and get data that is better resolved before plotting this image? Any suggestions are very welcome!
|