Small problem using the Batch Preprocessing script

[helgeras]

Hi,
I just bought PixInsight, and am new to astrophotography too, so the problems i have probably is caused by myself. Anyway, I would like to understand what I am doing wrong...

I have
- 10 x 120 sec SII filter lights
- 10 x 120 sec Ha filter lights
- 10 x 120 sec OIII filter lights and
- 10 x 120 sec darks
I have no flats, nor any bias frames

These I loaded into the batch preprocessor script, ensured that I had 3 sections of lights (one with 10 pictures each for each filter), and darks listed.
I then ran the script with all the options as they were set by default.
The result looks fine except that there are some spots of red and green pixels.

You can see an example of this in the little attached picture to this post, or you can check out the resulting fits file here if you want: https://dl.dropbox.com/u/7714699/M27.fits
If you zoom in and look around a bit then you will spot them easily.

My guess is that it is hotpixels that hasn't been corrected correctly.
My guess is that the star alignment step that I assume is done somewhere in the script moves the hotpixels, so that I get a group of them instead of just one.

What am I doing wrong?

Any help would be appreciated

Regards, Helge

[topboxman]

1. What kind of camera? Is it mono or color?
2. 120 seconds is pretty short for narrow band images. Usually a minimum of 15 minutes is required to gather good data for NB images.

Peter

[helgeras]

It is a mono camera, I took 10 frames with each of the 3 filters.
I know its way too little exposure time, but tell that to the clouds

[topboxman]

OK. I think I misunderstood your original post. I just downloaded your M27.fit and it looks great despite only 2 minutes sub-exposures. I do see some red/green/blue streaks.

Try set "Hot pixel removal" to 2 in Star Alignment.

If you didn't dither, dithering would have likely removed hot pixels during Star Alignment or Image Integration.

Peter

[helgeras]

Hi Peter, thanks for replying. I tried to set hotpixel to 2 and it helped a lot, but the streaks are still visible (but barely).
Where do I set up the dithering?

Regards,
Helge

[topboxman]

Dithering is done during capturing of images, not processing. You need software to do this. I use Nebulosity to capture and PHD for autoguiding. I link both software for Dithering. After Nebulosity is done capturing the image, it tells PHD to start dithering and when PHD is done dithering, it tells Nebulosity to start capturing next sub.

Maxim DL should also have Dithering feature as well.

I have Atik 460EX mono camera which uses Sony ICX694 CCD. Since Sony CCDs have very little to no noise, I don't use dark subtraction. I use Nebulosity's Bad Pixel Mapping which really helps remove "Bad" pixels. Bad pixels can be both hot, cold and/or dead pixels. If your camera uses Kodak CCD, then it may be better to use dark subtraction since they normally have more noise than Sony CCDs.

I am sure other software can use Bad Pixel Mapping but I do not know which one.

Peter

[helgeras]

Do you know if PixInsight has something compared to Nebulosity's Bad pixel mapping algorithm?

I am using PHD, and will read up on the dithering setup. Btw, I use the same camera as you

Regards,
Helge

[georg.viehoever]

I think the CosmeticCorrection process does what you need http://pixinsight.com/forum/index.php?topic=3483.msg32172#msg32172
Georg

[topboxman]

Helge,

What software do you use for capturing? PHD may not work with other capture software for dithering. I have read someone wrote some kind of plugin to link PHD with MaximDL for dithering. It's called MaxPHD.

http://www.skymonsters.net/software.php

If you use Nebulosity, then you won't have problems dithering with PHD. Also, try use Bad Pixel Mapping with Nebulosity.

Peter

[pfile]

OK. I think I misunderstood your original post. I just downloaded your M27.fit and it looks great despite only 2 minutes sub-exposures. I do see some red/green/blue streaks.

Try set "Hot pixel removal" to 2 in Star Alignment.

If you didn't dither, dithering would have likely removed hot pixels during Star Alignment or Image Integration.

Peter

does hot pixel removal in StarAlignment actually remove hot pixels in the image? the documentation implies it's only ignoring hot pixels while doing the star detection.

[topboxman]

I think you are correct. One time I was running Star Alignment and then Image Integration and resulted an ugly image because there were star trails and it must have picked a lot of hot pixels so I selected "2" for "Hot pixel removal" in Star Alignment and it solved the problem for Image Integration.

At the time, I was comparing the results using PixInsight calibration including dark subtraction with Nebulosity's Bad Pixel Mapping. I find Bad Pixel Mapping did a better job of handling bad/hot pixels. I see PixInsight Defect Map tool but I do not know how to use it and I do not know if it's similar to Nebulosity's Bad Pixel Mapping.

So my case is different than Helge. Now you mention it, I find it strange that setting "Hot pixel removal" to 2 helped Helge.

Peter

[helgeras]

setting the hot pixel removal to 2 made the hot pixels  much less noticeable, but they still were there.

However, I just found that if I click off "Optimize dark frames" in the script, then the pixels disappear.
I did it because the tutorial video on Harry's Astro Shed suggested it.
Not sure what that option does, but it worked

Helge

[topboxman]

That's great news.

I avoid dark subtraction mainly because of low noise Sony CCD chip and I was afraid that it may actually inject more noise. Bad Pixel Mapping does a nice job of removing hot pixels. Also dithering images helps as well. If there are still some hot pixels left after calibration, image integration may remove remaining hot pixels if the images were dithered.

Peter