Re: The promised format

[Herb Johnson <hjohnson@pluto.njcc.com>]

As Tom has noted, my Tech Note #71 includes discussion of the non-imaging
columns that occured a few years ago. It also provides my analysis of
some early dark images from the Mark IV, and some simple programs
to look at rows and columns of the Mark IV images. One can perform
a histogram analysis by rows or columns, avoiding obvious camera
and non-imaging effects, and get a better idea of actual imaging
performance. The analysis that I and others did at the time also
reveals some effects with the first few ROWS of the image. 
Also, there are apparently some effects along the edges of some images
and darks. Finally, one should keep in mind that due to various optical
effects, the corners of all images (but not darks) will have lower
background counts due to vignetting, coma, etc. (Tech Note #72
discusses coma and saturation using similar methods to TN #71 and
may be informative.)

For the
most part, these issues have been back-burner items for some time,
although I and others have brought them up on many occasions. The
prime purpose of my Tech Note was to provide that whenever that discussion
became "front burner", that prior discussion and results would not
be lost. I'm glad Tom found this TN useful in just that way.

Regarding some global methods for finding abberant pixels; I 
found that a relatively straightforward method was to look at
a row or column at a time, compute a distribution, and to reject
values three sigma from the average obtained. Generally this
resulted in the loss of a few to several pixels out of 2000,
and a much lower standard deviation, without any substantial
change to the average value. The TN details this method and the
results on the darks for Tom's early mark IV work. One consequence
of this method is to provide a way to identify a criterion for
abberant pixels (pardon my spelling, I'm composing this note online
to save time.) 

When time
permits, I will use similar methods for more recent darks and
images. That was the second reason to "post" this tech note,
so that I would not have to explain my methods again.

I believe that more recent darks, produced by a camera at lower
temperature, have only a slightly lower dark background. Tom
has suggested the standard deviation is lower as well I believe.
My recollection of my own viewing of recent images from Tom suggests
that some but not all of the "edge effects" of his early images are
reduced in his recent images. The non-imaging column format has not
changed to my casual observations, and Tom has said he's not worked
on his ROM-based imaging format to change the format.

As time permits I will look at disk #15 and #16 images, but the
election and Thanksgiving took my time up in November. In any
case the TN lays all of this out in preparation, and accredits
findings and methods appropriately.

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
     classic S-100 computers restoration & parts as "Dr. S-100"
   rebuilder/reseller of classic Macs for your computing pleasure
          and senior engineer at Astro Imaging Systems

On Wed, 6 Dec 2000, Tom Droege wrote:

> OK, Herb Johnson has captured my previous note on what I am writing, and 
> put it in TN-71.  This is repeated below.  So you don't have to wait until 
> tomorrow.
> 
[snip]
> 
> Note from Herb: a number of TASS members have reported that the first 6 
> columns and the last 2 or 3 columns of the images on CD #5 are clearly NOT 
> imaging columns.
> 
> OK, this is where we stand.  I will not turn my attention to writing up 
> everything.  Perhaps someone can detect the problem.
> 
> Tom Droege