RE: mark iv compression

["Gamble, David" <dgamble@anglogold.com.au>]

We have been using enhanced compression wavelet for airphotos and
geophysical images. Compression ratios are very high, ranging from 1:25 to
1:50. This is public domain software available from
http://www.ermapper.com/product/ermapper6/ecw/index.htm. ERmapper also have
a very powerful image web server that can distribute terra byte images over
the Internet in real time. I am not sure how lossy this would be for
photometric purposes but given the very high compression ratios it may be
worth a look.

-----Original Message-----
From: hjohnson@pluto.njcc.com [mailto:hjohnson@pluto.njcc.com]
Sent: Thursday, June 08, 2000 5:46 AM
To: tass@listserv.wwa.com
Subject: Re: mark iv compression


On Tue, 06 Jun 2000 17:16:23 -0700, Chris Albertson <calbertson@logicon.com>
wrote:
*>
*>One thing I thought of that I'll ssuggest we concider:  I seems that we
all
*>want to keep the raw image data, Tom is keeping it just as it came off
*>the camera and Arne has suggested after 10X compression.  After reading
*>Jures tech note I thinkg we should at least modify the header to include
*>corrected WCS (ra, dec, image scale and rotation and maybe more) before
*>the data is archived.
*>> 
*>  Chris Albertson

Some of you may recall I have a Tech Note on compression, written before
Mark IV data was available. While I've updated the references in it,
I have a bit more information to include which I will add shortly.

I've reviewed our recent correspondence on compression: I don't see offhand
any references to compression tests with current Mark IV data. I may have
missed
something, I'd appreciate any info sent to me so I can add it to my
Tech Note. (My review was cursory, no offense if I missed some recent work.)

But my impression is that NOW is the time to test the "new"
uncoma-ed Mark IV image data for compressibility, if one or more sets
of reduction tools and/or programs are available for A - B testing.
That is, results from compressed vs. noncompressed images. The Tech
Note I believe will suggest the kinds of testing done previously on Mark
III.
While it suggests methods, I would expect different results.

Tom's recent suggestion was *not* to keep the entire image, but to keep
the pixels measured ONLY. In effect, compressing the image "array" to only
those components which contribute to starlists, if I read his suggestion
correctly. One could do the arithmetic, but I would guess that would be
on the order of 10:1 compression. That too should be tested.

By the way, the local computer superstores have recently offered 20GB
or 30GB hard drives for $99. I will again suggest the least effort
kind of "archiving" may well be to fill up one of these drives and to
put it on the shelf, perhaps with a $10 '486 computer wrapped around it.
There will be no shortage of large, cheap hard drives and cheaper, previous-
generation computers.

Herb Johnson

Herbert R. Johnson              http://pluto.njcc.com/~hjohnson
hjohnson@pluto.njcc.com         voice 609-771-1503, New Jersey USA
             amateur astronomer and astro-tour guide
           S-100 computer parts, manuals as "Dr. S-100"
    reseller of 68K Macs & accessories for your computing pleasure


**********************************************************************
Please ensure email address and references to
Acacia Resources Limited are changed to
AngloGold Australasia Limited effective immediately.

Email addresses are changed by replacing acacia.com.au
with anglogold.com.au

**********************************************************************