RE: [APML] New Image

["Brian Lula" <brianl@polytecpi.com>]

HI Jeff,

Thanks! 

For the blooms, I remove them in Photoshop with the rubber stamp and smudge tool in the 
LRGB image using immediately adjoining image information to the bloom and changing 
brush sizes to eliminate any star pickup. About a year ago I tried creating artificial 
diffraction spikes but they didn't look very natural and scrapped the idea. This time I 
outlined the "best loooking" diffraction spike of the target star, feathered it a few pixels 
including about 30% of the stars core and then copied it to the clipboard. I then rotated the 
image to the orientation ( there might be an easier way to do this) I needed to paste and 
position the spike to match it up with the star's core and it's opposite spike. If the 
surrounding field is uniform very little else needs to be done but if it complex in color and 
brightness I sometimes use the RGB data ( with much reduced blooming) to create an 
accurate background and then paste that in a similar manner.

This image though got me thinking about the bloom direction. I have always lined up my 
CCD camera so that the blooms ran in the direction of the diffraction spikes. After working 
this image I wondered if it would be easier to just work on removing blooms with out the 
added complexity of painting in the spikes by changing the angle of the camera so that the 
blooms were in between the diffraction spikes. I guess it may be safe to say that if you 
don't want to process the blooms then blooms in the spike direction is preferable but if you 
are going to rework the blooms then position the camera so that blooms are in between the 
diffraction spikes. Real short exposures can eliminate blooms but at some expense in SN. 

For the smoothing I didn't use any in the center regions of the image but did use a 0.8 pixel 
radius gaussian blur on the faint outer regions to reduce the noise. Another thing I would 
like to have done in retrospect is use Registar on the RGB combine to eliminate some of 
the smeared RGB star colors in the outer edges of the image field. I used MAXIM for the 
RGB combine (auto star matching method) but believe I had some image scale change 
during the long exposure needed to take this image due to changing  temperature. 

Hope this helps.

Best regards, Brian

On 4 Jul 02, at 0:04, Jeff Ball wrote:

> Brian,
> That is a wonderful image.  How do you process the blooms out and yet
> maintain the spikes?  Also, what did you use to smooth the background?
> So that's what frost spikes look like.  Thanks for the heads-up.  Best
> wishes.
> Jeff Ball
> 
> > 
> > The "artisticly" processed image is at:
> > 
> http://www.heavensgloryobservatory.com/Color_Jpegs/conenebFRGB01-6303Ewe
> b.jpg
> 
> The "raw" combined color/b&w image is at:
> 
> http://www.heavensgloryobservatory.com/Color_Jpegs/conenebFRGB04rawcombi
> ne.jpg
> 
> 
> 
> 
> --  APML Archives at <http://astro.umsystem.edu/apml/>  ---
>              Unsubscribe at <majordomo@seds.org>




Brian Lula
Polytec PI, Inc.
(508) 832-3456

--  APML Archives at <http://astro.umsystem.edu/apml/>  ---
             Unsubscribe at <majordomo@seds.org>