- This topic has 9 replies, 1 voice, and was last updated 11 years, 11 months ago by
.
-
Posts
-
Posted by D A Dunn at 22:17 on 2012 Sep 08
I am new to this topic and am experimenting with some 5 sec shots off a fixed camera (D80). I am making dark frames and aligning the images using Deep Sky Stacker. It seems to work quite well and I am having fun picking up the Messier objects. However, how do I deal with images taken on different nights and having different darkframes. Is it OK just to stack the already stacked frames. eg night one produces a stack of 10 images and night 2 does the same can I stack the two frames from the two nights’ work?David
Posted by Graham Relf at 15:59 on 2012 Sep 09
There should be no difficulty, provided that there is a significant overlap between the frames, so that most of the brightest stars in one frame are also visible in the others. The scale needs to be similar too – ie, the same focal length lens or optical set-up in the telescope for all frames.My own software works by looking at the patterns of connecting lines between the brightest stars (the 24 brightest, unless the user changes that number). They do not all have to match.Stacking is done into an accumulator image having 32 bits per channel, so that adding up brightnesses does not cause them all to saturate (reach the 16-bit limit, which is the depth of the original images). For deep sky work it is important to keep the 32-bit depth for subsequent processing, otherwise you are throwing away the possibility of pulling out the faintest detail from the background. So if the stacking software can take as input files having 32 bits per channel, you will not lose anything by stacking two or more already stacked results. That necessarily means saving the results as FITS files because 32-bit TIFF files (aimed for importing into Photoshop) use 32-bit floating point rather than integer values and so really have only 23 bits per channel resolution (allowing for wasted sign bit).If you cannot use 32-bit FITS files as input to the stacker you would be better off stacking all of the original 16-bit frames in one sequence, even if from different nights.
Posted by Callum Potter at 17:45 on 2012 Sep 09
Sorry if its stating the obvious, but you should be stacking calibrated frames – ie, frames corrected with darks, flats and biases (from the night the images were made).Then there should be no problem stacking from different nights.Good stacking software should align even if there is some misaligment of frames.I have tended to use IRIS which also works with 32 bit integer data – a point Graham makes well.Callum
Posted by Andrea Tasselli at 14:40 on 2012 Sep 10
Just to clarify further; IRIS does NOT use in any form or fashion 32-bit deep images nor it is capable to read them. It only recognizes 8 and 16-bit deep images and uses 16-bit *signed* FITS files (or PIC) up to 3 planes per image.IRIS (and all the software I know of) will happily align and stack calibrated and UN-calibrated frames, so it is up to the user decide what to do.As long as there are 3 stars shared between all the frames IRIS is more than capable to align and stack them. In fact they do not have to have been taken with the same hardware *at all*, as long as there are enough stars to perform trasformation (to a common scale), alignment (with possible morphing) and stacking.Andrea T.
Posted by Callum Potter at 16:57 on 2012 Sep 10
Hi Andrea,thanks for clarifying – my memory does not seem to be what it used to be!Callum
Posted by Graham Relf at 09:13 on 2012 Sep 23
A zip of the PowerPoint file from my talk yesterday (DSLR Observers Workshop) can be downloaded from britastro.org/computing/gr/WhyStack.zip.It’s about 17 Mbytes.
Posted by Roy Hughes at 17:06 on 2012 Sep 23
Graham,thanks for the PP presentation, I wasnt there so it has been a very good read. It prompted me to revisit posts earlier in this thread about 32 bit floating point TIFF files.In an earlier life (pre-retirement) I worked in the city writing back office systems where microscopic rates of interest were applied to astronomical sums of money (not mine!) and spent a lot of time investigating the accuracy (or otherwise) of floating point numbers.Floating point numbers are always saved Normalised (i.e. no leading zeroes) so any any loss of accuracy is at the low end of the number. The normal IEEE representation as used in the math co-processor in your PC has the general format [8 bit unsigned exponent, sign bit, 23 bit mantissa]. The exponent is implicitly scaled giving a range -127 to +127. The first binary digit of the mantissa is omitted (as by definition it will always be 1) and the next 23 bits stored. The 32 bit floating point TIFF format follows this standard (if the document issued by Adobe in 1992 is to be believed) and therefore the accuracy is 24 rather than 23 bits (7.225 decimal places).The final precision when holding integer numbers in floating point would vary according to the size of the number being represented, a fully saturated pixel would lose accuracy (but only at the 1 part in 16 million odd end, where it really would not matter) but not magnitude. At the other end of the scale a single 1 bit value would have 23 spurious binary places of accuracy.I can see that the software manipulating the numbers could need 32 bit integer working registers but 32 bit TIFF files should work quite acceptably for storing images.Have I missed something here?. Over to you.Roy
Posted by D A Dunn at 14:05 on 2012 Sep 30
Graham,Thanks for the link to your presentation. Just got back from 2 weeks in Provence with very dark skies, looking forward to seeing what the images are like that I took.RegardsDavid
- You must be logged in to reply to this topic.