|
Go
![]() |
New
![]() |
Find
![]() |
Notify
![]() |
Tools
![]() |
Reply
![]() |
|
|
Poor Seeing |
Has anything changed with set background tool in the latest version? I can not achieve a neutral background with the latest version. It used to work flawlessly in the past.
Regards, |
||
|
|
Poor Seeing |
CCDStack is a great product - and superb at aligning images, data rejection, bloom removal, pixel and file maths, and combiening images.
However, it's ability to remove a background gradient is average. As far as I know, only one program does this really well - IRIS by Christen Buill. And it's free! It works by automatically adding thousands of sample points that avoid stars and nebula. It then fits a polynomial to the background sky - you can select the order of the polynomial - from linear to high order. You can also create a mask from the current image to help the sample points avoid nebula / galaxies. But there is a snag - it can only cope with a limited dynamic range of -32768 to 32767. So I do all my initial processing in CCDStack, but use IRIS for background gradient removal and noise reduction (try adaptive gaussian filter) and then back to CCDStack for the DDP visualisation. Stan - any chance of a similar system for background gradient removal in CCDStack in the future? That would be fantastic! John |
|||
|
|
Poor Seeing |
In the past, I was able to neutralize the background with a simple click and selection. Using the latest version, I see bias in color. Sometimes red and other times blue. I am just wondering if something has changed?
Regards, |
|||
|
|
Orbiting around Earth |
"Set Background" does not flatten the background or correct changing hues. It simply ascertains the nominal background ADU and applies color offsets to neutralize the color of that ADU. If there are differential color gradients then “set background” will balance only the area selected and other background areas may exhibit residual color errors. There are a few ways to deal with this:
• Select the brightest background area (a background gradient will have brighter and darker areas). • Check the “de-saturate” box. • Split the colors (if not already separate) and flatten each color, then recombine and set background. Stan |
|||
|
|
Good Seeing |
I am also having trouble with create color. When I do the set background fuction the whole image becomes bright red and blue/green. It is not just shaded red and blue/green it is like saturated red and blue/green. Help.
George |
|||
|
|
Orbiting around Earth |
could you send me that image?
(stan_ccd at yahoo dot com) Stan |
|||
|
|
Good Seeing |
Stan,
I sent you an e-mail with the color image in 32bit FITS format. Thanks for your help. George |
|||
|
|
Good Seeing |
Stan,
The e-mail was bounced by yahoo.com because it was too large 17.9 MB. George |
|||
|
|
Good Seeing |
Stan,
I uploaded the file to my iDisk Public folder. I sent you an e-mail with a link to the file. Thanks, George |
|||
|
|
Poor Seeing |
I have re-processed an old image with the latest version of CCD. Then I compared the histogram of the old image with new image after appling the set background tool. Next, I open them in PS. The background was correctly neutralized with the older version. The new image was shifted in bias. None of the colors were aligned correcrtly. The Gamma is acting strange. After using autoscale, I notice it is set to 0.95. I just updated my CCD inspector to the latest version. The new plug-in was added to CCDStack. I have been using this program for years without a problem.
Regards, Mike |
|||
|
|
Orbiting around Earth |
Hi Mike,
There have not been any significant changes in "set background" for some time. But there has been a recent change in DDP Auto-Scale, whereby it is more aggressive (applies a gamma stretch and a lower DDP value). The new scaling can definitely reveal background errors that were muted in the earlier autoscale. Try changing the gamma to 1.0 and increasing the DDP value. If you compared scaled images (e.g. JPEG) then it is not surprising that the histograms would be different. But if you were comparing linear FITS then a noticeable difference is unexpected. In which case, please contact me privately so that I can obtain those images. Thanks, Stan |
|||
|
| Powered by Eve Community |
| Please Wait. Your request is being processed... |
|

