SharpCap Forums

I recently had a clear night but with some intermittent wind. The lunar captures did not stack well. This document describes how the data was cleaned up using PIPP, thus enabling Autostakkert to produce clean stacks.



[Edit: I forgot to mention, this technique can be useful if mosaic panels will not stitch seamlessly. Pre-processing with PIPP before stacking can help the stitching software achieve a better join].

Dave

Excellent, Dave!

Thanks Brian. I will reserve judgement until someone tests the document out on some affected data.

Dave

Dave,

It looks like you chose an area of interest in PIPP that did not include the portions affected by the wind judder (blacked out in AutoStakkert). What would the result have looked like had you included the whole first frame in the PIPP area of interest (or had not set an area of interest)?

Thanks,
Don

Don

What would the result have looked like had you included the whole first frame in the PIPP area of interest

If the whole frame is the area of interest then many frames would be discarded by PIPP. As soon as any part of the AOI is outside the frame then PIPP discards the frame.

or had not set an area of interest

In this case PIPP would retain all frames which completely contained the Anchor Feature Box (AFB). the stack would have ragged edges and need to be cropped.

I did have a few tries with the AOI until I had a satisfactory result.

This page from the PIPP manual is worth a look at https://sites.google.com/site/astropipp ... e/example8. My situation was very much like the first video - sometimse most of the target area was 'off frame'.

I think that to best understand this, it is worth capturing some lunar videos on a windy night and experimenting with the software.

It is worth capturing many frames to increase the chances of acquiring the target area. I now have a camera with can capture MONO12 at 120fps over USB3 for a 640x480 capture area.

I hope you found the document useful.

Dave

Hi Dave,

Yes, I did find the document useful.

I usually use the AutoStakkert framing sliders to initially set a smaller ROI to analyze and stack, partly to minimize framing issues caused either by wind or imperfect tracking and partly to speed up the stacking process. Even then, I sometimes still end up with either some partial frames or frames with some other anomaly toward the front of the quality-ranked stack, which would then be stacked. It would be nice if there was some automated way to either prevent this or would show you the largest ROI (or AOI in PIPP) that would have complete frames instead of having to iterate to find an acceptable solution.

Don

Yes, I did find the document useful

Thanks Don. I am always left wondering if my authoring efforts help anyone.

Lunar imaging is a lottery with wind, thin cloud and atmospherics. Over the last couple of years I have collected around 1.5Tb of lunar data - only the processing shows if the data collected was any good. These days, if organised, I would expect to capture around 100Gb in a 1 hour session. The equipment I use is:

• Celestron C8 SCT

• Skywatcher 127 MAK (gives much better results than the price suggests)

• QHY5LII-M camera

• Celestron Skyris 618M (recent addition, no images yet)

• Celestron 2x Barlow

• reducers, 0.63x for the C8 and 0.5x for the 127 MAK

• IR 685 pass filter

• Wratten #25 red filter

• IR cut filter

I consider these to be amongst my better efforts:

• https://www.astrobin.com/full/uj3ns8/B/?nc=user

• https://www.astrobin.com/full/umry65/B/?nc=user

• https://www.astrobin.com/full/5lvi9x/0/?nc=user

• https://www.astrobin.com/full/topkd9/0/?nc=user

Dave

Wow, you do a lot of lunar! Nice shots - they look like they were from a lunar orbiter!

Thanks Don, all captured at the bottom of the garden.

I found the solution that I needed if only a limited number of frames are affected. In AutoStakkert!, after doing the analyze frames, hit the enter key while at the frame you want to delete. This imparts a red coloring to the image size box. Select analyze again and the frame will be deleted from the new set of analyzed frames.

Don