Xref: utzoo comp.graphics:2412 comp.misc:2324
Path: utzoo!mnetor!uunet!husc6!bloom-beacon!gatech!gitpyr!ccastks
From: ccastks@pyr.gatech.EDU (Ken Seefried iii)
Newsgroups: comp.graphics,comp.misc
Subject: Re: Image compression using fractals?
Message-ID: <5703@pyr.gatech.EDU>
Date: 4 May 88 04:24:19 GMT
References: <220@turbo.RAY.COM>
Reply-To: ccastks@pyr.UUCP (Ken Seefried iii)
Organization: Georgia Institute of Technology, Atlanta
Lines: 36

In article <220@turbo.RAY.COM> Robin@turbo.RAY.COM ( Dr. R.P. Alston) writes:
>
>I recently read an article in Byte magazine concerning image compression
>using fractals and obtaining ratios up to and greater than 10,000 to 1.
>Can anyone recommend some more reading on this, or is there anyone on the
>net involved in this research? I am much interested in looking into whether
>this is going to be a real and feasable system to image compression with
>special hardware in the short or long term future. Thanx for any and
>all help. Mail first please.
>
>-- 
> -------------------------------------------------
># Whats worse than two MA drivers on a freeway?    #      Dr. Robin the REAL
># Answer: One Toronto driver                       #      SuperUser Atilla!
> -------------------------------------------------	  (617)-460-8225
>	Robin@turbo.ray.com		.....!rayssd!turbo!Robin

Many moons ago, I worked briefly in Micheal Barnsleys lab and took his first
class in fractal.  The course was taught from class notes, which he is evolvin
into a textbook on fractals, or to be more precice, chaotic dynamical systems.
The book is not yet published, but from the material that I was exposed to,
it will be very good.

I know of two other references that you can check: the original book by
Benoit Mandelbrot (i forget the name) and a very terse (read lots of math)
book called "An Introduction to Chaotic Dynamical Systems" by Robert L.
Devany.  The Devany book lays down a lot of fundamental stuff.

As far as being practical, i would say yes.  There was talk for quite a 
while of some big industry funding for an IFS chip to do the compression.
In any case, the compression figures of 10,000 to 1 are true.  The particular
application that isgenerating most interest is compression of satellite 
photos.

	ken seefried iii
	ken@gatech.edu