NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/sources/wanted/5132

Text File | 1992-11-23 | 1.8 KB | 47 lines

Xref: sparky comp.sources.wanted:5132 comp.graphics:12240
Newsgroups: comp.sources.wanted,comp.graphics
Path: sparky!uunet!mcsun!sunic!ericom!eos.ericsson.se!etxmesa
From: [e-mail] (Michael Salmon)
Subject: Re: Postscript optimization
Message-ID: <[e-mail]>
Keywords: postscript,bitmap,compression
Sender: [e-mail]
Nntp-Posting-Host: eos6c02.ericsson.se
Reply-To: [e-mail] (Michael Salmon)
Organization: Ericsson Telecom AB
References:  <[e-mail]>
Date: Mon, 23 Nov 1992 11:19:26 GMT
Lines: 31

In article <[e-mail]>, steve@denali (steve) writes:
|> Does anyone know a way to reduce the size of huge postscript image
|> files?  Our application produces only bitmaps for shaded output which
|> can be several megabytes in size.  Print time can be hours, with the
|> biggest delay being time to send through the parallel port.
|>
|> I'd like to optimize the file, perhaps by finding areas of nearly the
|> same color or intensity and polygonalizing them.

I've been thinking of something similar, the idea is quite old but to
my knowledge it hasn't been used for PostScript. We increase the number
of pixels to a power of 2 and equal in each direction i.e. 1152x900
becomes 2048x2048, unused pixels are defined as white. For each group
of 4 pixels we create a higher order pixel that has the most common
colour or white if there is no common colour. We repeat this process
until we have a single super pixel. The page can then be drawn using a
series of rectangles that are filled, I think that white can be ignored
at all but the lowest level. By defining a prodecure to do the work it
should be quite easy to encode each rectangle with just a few
characters.

--

Michael Salmon

#include    <standard.disclaimer>
#include    <witty.saying>
#include    <fancy.pseudo.graphics>

Ericsson Telecom AB
Stockholm