spectro-aras.com

Hi,

I am a bit confused about the determination of the cosmetic files in ISIS. I have been playing with various thresholds and dark frame integration times using my Atik 314L camera. I find that there are differing numbers of hot pixels for the different dark frames taken at different integrations, so it does appear to me that I should create a cosmetic file for each dark frame time. I also noticed differing numbers of hot pixels for dark frames taken at different temperatures. But I also noticed that there were differing numbers of hot pixels for differing thresholds. For example, at thresholds between 300 and 100 in a 60 second dark frame,, the numbers of hot pixels ranges from 82 (threshold 1000) to 184(threshold 300) hot pixels. At threshold 250, the number jumps to 60988. A threshold of 300 appears to be the least threshold value I should be using regardless of camera temperature.

So my question is how can I tell which threshold I should be using? Is there a way to determine the optimum threshold for dark frames of various temperatures, cameras, and integration times, I use an Atik 314L camera on my Alpy and an SBIG ST8xe on my LhiresIII.

Cheers,

Keith Graham

Hi Keith,

I have not used this feature but as nobody has answered, I will add some thoughts and if they are wrong perhaps someone else will jump in.

As I understand it the cosmetic file is to identify hot pixels in the dark which are either already saturated or might saturate in the image and so will not be corrected properly by dark subtraction. these are then corrected in a way similar to cosmic ray hits using an average of the surrounding pixels.
If this is the case I think you should set the limit as high as possible (equal to the difference between the maximum count you normally run to in the image and the count at which your camera starts to saturate.) The main thing is to only include hot pixels which are there in all exposures (ie those that are still high in a stack of darks, not ones that come and go (I have noticed quite a few of these with my 314L+ which has low average noise but seems to have a long tail in the distribution) so you need a good stack of darks to identify them.

I believe also there is a trend in imaging to use cosmetic files as a substitute for dark frames these days when dark noise is so low with modern cameras. In this case I think you would need to use a lower threshold though.

Note these are only my ideas and are not based on any experience so could be totally wrong!

Cheers
Robin

Hi,

Sorry for the late answer. The note of Robin describe very well the situation.

Threshold adjustement of cosmetic tool is subjective.

There are not very precise rules, but a little experience. With a camera Atik314L + at -5 to -10 degrees
selected threshold level such that the number of hot spots (not necessarily saturated), is between 200 and 300.

For Atik460EX year, I let ISIS select 500-600 hot pixels, typically.

These values ​​are a good compromise for corrrect treatment.

Find the value by trial and error.

Christian

Thank you Robin and Christian for your replies. Based on your comments, I think I now understand what thresholds I need to apply at the various time integrations at 0 degrees in order to obtain a hot spot value of 200-300.

Cheers,

Keith