360 degree panorama of night sky

Discussion forum for Tawbaware's Star Tracer and Image Stacker software
Post Reply
Posts: 44
Joined: Sat Jan 17, 2009 4:48 am
Location: Newcastle, Australia

360 degree panorama of night sky

Post by Growing » Sun Jul 11, 2010 9:28 pm

NASA - Astronomy Picture of the Day - July 11 2010

It is a 4 hour 360degree panorama of the night sky. Unfortunately it is displayed in a highly distorted projection.

Assuming one wanted to reproduce this image in Star Tracer (with a more interesting foreground), what would be the best work-flow?

1) Just do one 360 pano set with 30s exposures, stitch it up into a cylindrical projection, and then extend the star trails 2 hours forward and 2 hours backwards (so you don't get gaps at the horizon)?

2) Or should you aim to take a series of 360 panos over the period of 4 hours, merge them, and then use Star Tracer to fill in the gaps?

3) Or should you take a series of 30s exposures at one position (say 20 over 10 minutes), then a series of 30s exposures at the next position, etc, until you have completed the 360 view. You then merge each position together (easy since alignment will not be a problem), then stitch each position together to form a cylindrical projection. Then use Star Tracer to fill in the gaps between exposures (a few seconds of gap). Then use Star Tracer again to extend the 10 minute trails 2 hours forward and 2 hours backwards to complete the image.

I suspect the 2nd method is best technically, but I'm almost certain I would have a lot problems stitching the sky accurately enough to ensure that the star trails will merge smoothly.

I have suggested 30s exposures because that is the longest non-bulb exposure on a DSLR, and hot pixels seem to become too much of a problem with longer exposures.

Any thoughts on the logistics of reproducing this image?


Posts: 3649
Joined: Fri Jun 20, 2003 8:55 pm
Location: USA

Re: 360 degree panorama of night sky

Post by maxlyons » Mon Jul 12, 2010 9:04 am


I think that all three of those approaches are viable. I'm still learning how best to use Star Tracer myself, so I don't think I have a definitive answer. However, here are some thoughts on the three approaches you suggest:

Approach 1. This should work, but it may be difficult to coax Star Tracer into accurately determining the trajectories of the stars if the trails are only 30 seconds long. You'd need to be extremely accurate with the placement of the star lines, and/or you'd need to set a large number of star lines.

Approach 2. My concern here would be ensuring that you get the panos merged accurately. My tripod isn't accurate enough for me to assume that I can precisely repeat the positioning of the camera, so I'd need to use PTAssembler to align the completed panos before merging. And, if that alignment isn't done 100% accurately, then the trails won't appear to "connect the dots" between trail segments.

Approach 3. I think this is my preferred approach. It ensures that the trails are long enough to allow them to be extended with little difficulty, and you shouldn't have any problems with positioning images.


Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest